Here's a question that should be easy to answer: how ready is your marketing operations team for AI? Not "are you using AI" - most teams are, whether they realize it or not. The platform features, the scoring enhancements, the content tools, the send-time optimizers. AI is already inside the stack. That's not the question. The question is whether you know what AI is doing in your environment, whether your data is in a state to support it, whether your team has the skills to manage it, whether there's any governance around it, and whether anyone can confidently say your organization is ready for what comes next - not just the AI you've adopted, but the AI your platform is about to ship, the AI your competitors are deploying, and the AI that regulators are about to start asking questions about. Most teams can't answer that. Not because they're behind - because nobody's asked. Nobody's assessed. Nobody's measured. There's no baseline. And without a baseline, every AI decision is a guess. The confidence gap There's a particular kind of confidence that's common in B2B marketing teams right now. The team is using AI tools. Campaigns are running. Content is being produced faster. The platform's AI features are active. Leadership has been briefed. Everything feels like progress. But underneath that confidence, there are questions nobody's sat down to answer. How mature is your data - not in theory, but right now? When was the last time someone checked whether the fields feeding your AI-powered scoring model are still accurate? Is your consent data current enough to withstand regulatory scrutiny? Are the AI features in your platform configured deliberately, or did they get activated during an upgrade and nobody reviewed them? Does your team know how to evaluate AI outputs, or are they trusting whatever the platform produces? Is there a process for detecting when AI-driven decisions start drifting? Is anyone monitoring whether the AI is actually improving results, or has "we have AI" become the result in itself? The gap between feeling ready and being ready is where the risk lives. And most organizations can't measure that gap because they've never tried. Why self-assessment matters now Three things are converging that make self-assessment urgent rather than optional. Regulatory pressure is arriving. The EU AI Act's main provisions take effect in August 2026. Transparency and documentation requirements apply broadly - not just to high-risk AI systems. Any organization whose automated systems affect EU residents needs to be able to explain what those systems do, what data they use, and how decisions are made. That explanation requires knowing what's running in your environment - which requires an assessment. AI adoption is accelerating without governance keeping pace. Every platform is shipping new AI capabilities every quarter. Teams are activating features faster than they're governing them. The gap between what AI is doing inside the platform and what anyone can explain about it grows with every upgrade cycle. An assessment catches that gap before it becomes a liability. The competitive landscape is splitting. The organizations that understand their AI maturity - where they're strong, where the gaps are, and what to prioritize - are making better decisions about what to adopt, what to defer, and where to invest. The ones operating on assumption are adopting everything, governing nothing, and hoping the results justify the spend. The split between these two groups is getting wider. What an honest assessment reveals Most teams that go through a structured AI readiness assessment are surprised by what they find. Not because the findings are catastrophic - because the picture is uneven in ways they didn't expect. Data readiness is almost always lower than assumed. The team thinks the data is clean because the dashboards look fine. The assessment reveals consent records that haven't been reconciled in two years, scoring models calibrated to a buyer profile that's shifted, and enrichment sources nobody's reviewed since the contract was signed. AI is making decisions on all of it. Governance is almost always more fragmented than it appears. There's a policy somewhere. But the assessment reveals that nobody can produce a complete list of active AI features, nobody owns the AI layer as a distinct operational responsibility, and there's no process for detecting when AI-driven decisions drift. The governance exists in principle but not in practice. Team capability varies dramatically. Some team members are confident and skilled with AI tools. Others are activating features they don't fully understand because nobody provided training. The assessment reveals whether the team's AI capability is broad enough to support the AI footprint they're operating - or whether a few individuals are carrying the entire AI competency while the rest of the team works around it. Alignment across functions is weaker than expected. Marketing, sales, IT, legal, and compliance each have a partial view of AI in the organization. The assessment reveals whether those views are consistent - and they almost never are. Marketing thinks governance is handled. Legal thinks marketing is handling it. IT thinks the platform vendor is handling it. Nobody is handling it. Knowing where you stand changes the conversation The value of an assessment isn't the score. It's what the score makes possible. A team that knows its data readiness is strong but its governance is weak can prioritize governance without questioning its data investment. A team that knows its AI adoption is ahead of its team's capability can invest in training before the gap creates problems. A team that knows it's ahead of its industry peers can move faster with confidence. A team that knows it's behind can make a case for investment with evidence instead of anxiety. Without the assessment, every conversation about AI readiness is based on feeling - and feelings are unreliable. "I think we're in good shape" isn't a strategy. "We assessed at 7/10 on adoption, 4/10 on governance, and 5/10 on data readiness — here's where we need to invest" is a strategy. The baseline turns vague concern into specific action. That's what most teams are missing. Take the assessment Sojourn Solutions is building an industry benchmark report on AI adoption, governance, and operational readiness within marketing operations in 2026. As part of it, we've built an assessment that gives you a clear snapshot of where your organization currently stands. It takes around 7 minutes. You get your results immediately. And your data contributes to an industry-wide picture of where MOPs teams actually are with AI - not where vendors say they should be. Take the AI Benchmark Assessment → The teams that know where they stand will make better decisions than the ones that don't. The assessment is the starting point.

Ask a marketer from 15 years ago what they spent most of their time on and they'd say something like: understanding the customer. Researching what they need. Figuring out how to reach them. Writing something that would make them pay attention. The work was about people - understanding their motivations, their hesitations, their decision-making process. Ask a marketer today what they spend most of their time on and you'll get a very different answer. Configuring the MAP. Setting up workflows. Troubleshooting the CRM integration. Building segments based on field values. Checking why the sync broke. Learning the new feature the platform shipped last week. Figuring out why the report shows different numbers depending on which dashboard you pull it from. The work shifted. Somewhere along the way, marketing became less about understanding the buyer and more about operating the machinery that's supposed to reach them. The platforms won. The people got lost. The platform ate the profession This didn't happen overnight. It happened gradually, one tool at a time. First the email platform arrived, and marketers learned to think in terms of open rates and click rates instead of whether the message actually resonated. Then the marketing automation platform arrived, and marketers learned to think in terms of workflows, triggers, and scoring models instead of whether the buyer's journey made sense. Then the CRM integration arrived, and marketers learned to think in terms of field mappings, sync errors, and lifecycle stages instead of whether sales and marketing were actually aligned on what a good customer looks like. Each tool solved a real problem. Each tool also created a new layer of operational complexity that required someone to manage it. And the person managing it was usually a marketer - someone who was hired to understand customers and gradually became an administrator of systems. The job title says marketing. The job description says platform management. The gap between the two gets wider every year. The skills that get hired for have changed Look at a marketing job posting from 2010 and you'd see requirements like: strong writing skills, understanding of customer psychology, experience with brand positioning, ability to develop messaging that resonates with target audiences. Look at a marketing job posting today and you'll see: experience with Marketo/HubSpot/Eloqua, proficiency in Salesforce, knowledge of HTML/CSS for email templates, experience with marketing attribution tools, familiarity with data management and segmentation. The hiring criteria shifted from understanding people to operating technology. The marketers getting hired today are selected for their ability to work inside platforms, not for their ability to understand the humans those platforms are supposed to reach. This isn't wrong, exactly. The platforms are complex and someone needs to run them. But the imbalance is real. Most marketing teams are over-indexed on people who can operate the machinery and under-indexed on people who can tell you whether the machinery is pointed at the right audience with the right message. The campaign gets built. The buyer gets forgotten. Watch how a campaign gets created in most B2B marketing teams. Someone decides a campaign is needed. A brief gets written - usually focused on the asset (the ebook, the webinar, the email) and the mechanics (the segment, the workflow, the follow-up sequence). The team builds it inside the platform, tests it, and sends it. At no point in that process does someone typically stop and ask: why would the buyer care about this? Not "does this hit our MQL target." Not "does this align with our content calendar." Why would a real person, sitting at their desk, dealing with their actual problems, stop what they're doing to engage with this? That question used to be the starting point. Now it's an afterthought - if it's thought about at all. The process is optimized for building and sending, not for relevance. The team measures whether the campaign went out, not whether it mattered to anyone who received it. The result is a steady stream of technically competent, operationally sound campaigns that nobody particularly wants to receive. The emails are well-formatted. The workflows fire correctly. The segments are clean. And the human on the other end deletes it in two seconds because nothing about it spoke to their actual situation. Data replaced intuition. That's not entirely a good thing. The data-driven marketing movement was supposed to make marketing more effective by grounding decisions in evidence instead of gut feel. And in many ways it has. We know more about buyer behavior than at any point in history. We can track every click, every page visit, every email open, every form submission. But somewhere along the way, the data became a substitute for understanding rather than a tool for deepening it. Teams started making decisions based on what the data said without asking what the data meant. Open rates went up - but did the message actually resonate, or did the subject line just trigger curiosity? Click rates improved - but did the content deliver value, or did the CTA create false urgency? MQLs hit target - but were those leads genuinely interested, or did the scoring model reward activity without distinguishing intent? Data tells you what happened. It doesn't tell you why. And understanding why - why the buyer engaged, why they hesitated, why they chose someone else - requires the kind of empathy, curiosity, and human understanding that no dashboard provides. The best marketers use data to validate and refine their understanding of the buyer. The worst marketers use data to replace that understanding entirely. The difference shows up in the work - one produces campaigns that feel like they were written for a real person, the other produces campaigns that feel like they were assembled by an algorithm. The pendulum needs to swing back This isn't an argument against marketing technology. The platforms are necessary. The data is valuable. The operational infrastructure that makes modern marketing possible at scale is a genuine achievement. But the balance is off. Too many marketing teams have become platform operators who occasionally think about the buyer, when they should be buyer experts who happen to operate platforms. The order matters - because the platform doesn't know who the buyer is. It processes data about them. Understanding them is a human job that no tool, no workflow, and no AI feature will ever fully replace. The teams that still start with the buyer - who is this person, what do they need, what are they worried about, what would actually help them - produce work that feels different. The emails get read. The content gets shared. The campaigns generate conversations, not just clicks. Not because the platform is better, but because someone took the time to understand the person before building the machine. Marketing technology made it possible to reach millions of people with precision and speed. It didn't make it easier to understand any one of them. That's still the hard part. And the teams that remember it's the hard part are the ones producing work that actually matters.

Something has shifted in B2B buying behavior and most marketing teams haven't caught up. The buyer who shows up on your website today is not the same buyer who showed up three years ago. Three years ago, they arrived early in their research process - looking for information, comparing options, trying to understand the category. They needed educating. They needed nurturing. They needed the whitepaper, the webinar, the email sequence that walked them through the problem and the solution. Today's buyer arrives pre-informed. They've already read about your category. They've already compared your product to the alternatives. They've asked an AI assistant to summarize the differences. They've read reviews on G2. They've asked their network on LinkedIn. By the time they land on your website or fill out a form, they're not at the beginning of their research. They're near the end. They don't need educating. They need confirming. And most B2B marketing strategies are still built for the buyer who needed educating. The information gap closed. Marketing didn't notice. B2B marketing strategy has been built for decades on one fundamental assumption: the buyer knows less than the seller. The seller has the expertise, the data, the insights. The buyer needs to be led through a journey - awareness, consideration, decision - with content mapped to each stage. That assumption held when information was scarce. When the only way to learn about a product category was to attend a conference, read a trade publication, or talk to a sales rep. When the vendor controlled the narrative because the vendor controlled the information. That world is gone. Information is abundant, accessible, and increasingly synthesized by AI tools that give the buyer a coherent answer in 30 seconds. The buyer no longer depends on the vendor for information. They depend on the vendor for validation - proof that what they've already learned is accurate, and evidence that this specific vendor can deliver. The marketing strategies that worked in the information-scarce world - gated ebooks that introduce basic concepts, nurture sequences that educate over weeks, webinars that explain the problem before presenting the solution - feel increasingly irrelevant to a buyer who figured all of that out before they engaged with you. What the pre-informed buyer actually wants The pre-informed buyer has specific needs that most B2B marketing isn't serving. They want proof, not education. They already understand the problem and the solution category. What they need is evidence that your company can deliver - case studies with specific results, customer references, implementation timelines, integration details. The content that matters isn't "what is marketing automation?" It's "how did a company like ours implement this and what happened?" They want specifics, not overviews. They've already read the overview. They asked AI for it this morning. What they can't get from AI is the specific detail about your product - how it handles their particular use case, what the implementation actually involves, what the limitations are. Honest, detailed, specific content is what separates your marketing from the AI-generated summary they've already read. They want speed, not sequences. A buyer who's near the end of their evaluation doesn't want to be enrolled in a six-email nurture sequence that starts with "the evolving landscape of..." They want to talk to someone, see the product, and get answers to their specific questions. The time between first engagement and sales conversation should be hours, not weeks. They want honesty, not positioning. Pre-informed buyers have already seen your competitors' messaging. They know what everyone claims. What cuts through is honesty - what your product does well, what it doesn't do, who it's built for, and who should probably look elsewhere. That kind of transparency is rare in B2B, which is exactly why it stands out. The nurture problem Most B2B nurture programmes were designed for a buyer who doesn't exist in the same numbers anymore. The traditional nurture assumes the buyer needs time and education. It sends content at intervals - a blog post this week, a case study next week, an ebook the week after - gradually building awareness and interest until the buyer is "ready" for a sales conversation. For a buyer who arrived pre-informed and ready to evaluate, this nurture isn't nurturing. It's delaying. Every email in the sequence that teaches them something they already know is an email that wastes their time and signals that your company doesn't understand where they are in their process. The nurture programme that works for today's buyer looks different. It's shorter. It focuses on proof and specifics rather than education. It adapts based on what the buyer has already done - if they've visited the pricing page and read a case study, they don't need an introductory email. They need a direct line to sales. The teams that still run education-heavy nurtures on every lead regardless of behavior are training their best prospects to be patient while their competitors are training theirs to be fast. The sales handoff is where most teams lose the pre-informed buyer Even when marketing adapts to the pre-informed buyer, the sales handoff often undoes the work. The buyer arrives having done extensive research. They fill out a form expecting a conversation with someone who understands their situation. Instead, they get a discovery call where the sales rep asks them to explain their business, their challenges, and their requirements from scratch - all information the buyer assumed the company already knew based on their engagement history. This is the moment the pre-informed buyer mentally downgrades the vendor. They did their homework. The vendor didn't. The sales rep is asking questions the marketing data should have already answered - what pages the buyer visited, what content they downloaded, what their company does, how big they are. The fix is operational: make sure the handoff includes context. When a lead is passed to sales, it should come with a complete engagement history, the account's firmographic data, and any intelligence about where they are in their evaluation. The sales rep should know more about the buyer than the buyer expects - not less. Adapt or get filtered out The pre-informed buyer isn't a trend that's going to reverse. AI tools are getting better at synthesizing information. Review platforms are getting more comprehensive. Peer networks are more connected. The amount of research a buyer can do before engaging with any vendor is only going to increase. Marketing teams that adapt will shift their content strategy from education to evidence. They'll shorten their nurtures. They'll speed up their handoffs. They'll invest in the kind of specific, honest, proof-heavy content that a pre-informed buyer actually needs - and they'll stop producing the introductory content that AI can generate better and faster than any marketing team. Marketing teams that don't adapt will keep building awareness campaigns for buyers who are already aware, nurture sequences for buyers who don't need nurturing, and educational content for buyers who already graduated. They'll wonder why engagement is declining, why form fills are dropping, and why the buyers who do engage seem impatient and unimpressed. The buyer got smarter. The question is whether your marketing catches up - or keeps talking to the buyer who used to exist.

There's a specific kind of marketing team that's always busy. Always launching something. Always behind on something else. The roadmap has 30 initiatives, 15 are in progress, 8 are overdue, and someone just added 3 more because the CEO saw something a competitor did. The team is exhausted. The work is spread across too many things to do any of them well. Campaigns launch half-finished because there wasn't time to QA properly. Content goes out without being reviewed because the next piece is already due. Reports get skipped because nobody has time to analyze what happened - they're too busy setting up what happens next. This team says yes to everything. And that's exactly why they're underperforming. Yes is the default. That's the problem. In most marketing organizations, saying yes is the path of least resistance. A request comes in from sales - "can we do a campaign for this segment?" Yes. The CEO wants a presence at a new event - yes. Product marketing needs email support for a launch - yes. A partner wants co-branded content - yes. Someone read an article about a new channel and wants to test it - yes. Each individual yes is reasonable. The campaign makes sense. The event could be valuable. The launch needs support. The partner relationship matters. Saying no to any one of them feels like obstruction - like the marketing team is being difficult instead of being helpful. But the aggregate of every yes is a team doing 20 things at 50% instead of 10 things at 100%. Resources get spread thinner with every commitment. Quality drops because there's not enough time to do the work properly. Impact drops because nothing gets the attention it needs to actually perform. The team isn't underperforming because it lacks talent or tools. It's underperforming because it never said no - and the workload grew until the quality of everything suffered equally. The cost of every yes is invisible When you say yes to a new initiative, the cost isn't just the time it takes to execute. It's the time it takes away from everything else. Every campaign that gets added to the roadmap pushes other campaigns back. Every ad hoc request that gets accepted delays the planned work. Every "quick project" that leadership drops in consumes the buffer that was supposed to protect the team's ability to do their core work well. These costs are invisible because they don't appear on a balance sheet. Nobody tracks the campaign that went out without proper QA because the team was building something else. Nobody measures the optimization that didn't happen because there was no time for analysis. Nobody counts the strategic work that got postponed indefinitely because the team was too busy executing requests. But the impact shows up. It shows up in campaigns that underperform because they were rushed. In content that doesn't convert because it was produced to hit a deadline, not to serve the buyer. In a team that's burning out because the workload never stops growing and nobody is authorized to push back. What saying no actually looks like Saying no doesn't mean being unhelpful. It means being honest about capacity and ruthless about prioritization. When a request comes in, the response isn't "no, we can't do that." It's "we can do that, but here's what it displaces." Making the trade-off visible is the most important thing a marketing leader can do - because most of the people making requests have no idea what the team is already working on. The CEO who asks for a presence at an event doesn't know the team is in the middle of a platform migration. The sales leader who wants a campaign for a new segment doesn't know the team is behind on three existing campaigns. The product manager who needs launch support doesn't know the team just lost a person and hasn't backfilled the role. When the trade-off is visible - "we can do the event, but we'll need to push the nurture redesign to next quarter" - the requestor can make an informed decision. Sometimes the event is more important. Sometimes it's not. But the decision is made with full information instead of blind optimism about the team's capacity. This requires marketing leadership to protect the team's bandwidth the same way engineering leadership protects sprint capacity. Nobody walks up to an engineering team and says "add this feature by Friday" without understanding the sprint. Marketing deserves the same discipline. Prioritization is the highest-value skill in marketing The teams that outperform aren't the ones doing the most. They're the ones doing the right things - and only the right things. That means having a clear framework for what gets done and what doesn't. Not a vague sense of priorities - an explicit, documented list that the team and its stakeholders agree on. These are the three things we're focused on this quarter. These are the requests we'll accept. These are the ones we'll defer. Here's why. When everything is a priority, nothing is. That's not a motivational poster line - it's the operating reality of most marketing teams. The quarterly plan has 15 "priorities" which means it has zero, because the team will spend the quarter reacting to whatever is loudest rather than executing against what matters most. The teams that say no have shorter priority lists. They commit to fewer things and execute them properly. Their campaigns are better because they had time to plan, build, test, and optimize. Their content is stronger because someone actually reviewed it. Their reporting is meaningful because someone had time to analyze it. The output looks like less. The impact is more. That trade-off is hard to sell internally - especially in organizations that measure marketing by volume of activity. But the teams that make the shift consistently outperform the ones that stay on the hamster wheel. How to build the muscle Saying no is a skill most marketing teams haven't practiced. It feels uncomfortable, especially in cultures where being busy is equated with being valuable. Building the muscle takes deliberate effort. Start with the intake process. Every request should go through a single channel - not direct messages, not hallway conversations, not emails to individual team members. A single intake point makes the total volume visible, which is the first step toward managing it. Evaluate every request against the quarterly priorities. If it aligns, it goes on the roadmap. If it doesn't, it gets logged for future consideration - but it doesn't get worked on now. The log is important because it shows the team isn't dismissing requests. It's sequencing them. Make capacity visible. Whether it's a kanban board, a sprint plan, or a simple shared document - the team's current workload should be visible to anyone who wants to add to it. When a stakeholder can see that the team is at capacity, the conversation shifts from "why won't you do this?" to "what should we deprioritize to make room?" Review quarterly. At the end of each quarter, look at what got done, what got deferred, and what the impact was. Over time, this builds evidence that focused execution outperforms scattered activity - and that evidence makes it easier to say no next quarter. The courage to be focused The marketing team that says yes to everything will always look busy. Dashboards will show activity. Content will ship. Campaigns will launch. The team will be exhausted and the results will be average across the board. The marketing team that says no will look less busy. Fewer things will ship. Some stakeholders will be frustrated that their request got deferred. But the things that do ship will be better - better planned, better executed, better measured, and more likely to produce the results that actually matter. The difference isn't talent. It's discipline. And the hardest part of that discipline is the first time someone says "can you do this?" and the answer is "not right now - here's why." That conversation is uncomfortable. It's also the beginning of a marketing team that actually delivers instead of one that just stays busy.

There's a test you can run to see how your organization thinks about marketing ops. Look at where the team sits in the org chart. Look at who they report to. Look at what they get asked to do on a daily basis. Then look at what gets said about them in leadership meetings. If the answers are "buried under demand gen," "a marketing manager who doesn't understand the platform," "build this email, fix this list, pull this report," and "nothing - they don't come up" - then your organization treats marketing ops as a support function. A service desk. The team that makes things go when someone else decides what should go. That's how most organizations treat MOPs. And it's costing them far more than they realize. The service desk trap When marketing ops is treated as a support function, the work becomes reactive. The team doesn't plan. They respond. Campaign requests arrive and get built. Data issues get flagged and get fixed. Reports get requested and get pulled. The team is permanently in execution mode, processing a queue of requests from other teams who decide what gets done and when. This feels productive. The team is busy. The queue is always full. Campaigns go out. Reports get delivered. From the outside, marketing ops looks like it's working. From the inside, the team is drowning. There's no time to audit the platform. No time to optimize scoring models. No time to document workflows. No time to evaluate whether the campaigns being requested are the right campaigns, or whether the data underneath them is reliable, or whether the reporting structure actually measures what matters. The team knows the platform better than anyone in the organization. They see every campaign, every data flow, every automation, every integration. They know where the problems are. They know what's broken, what's inefficient, and what's creating risk. But nobody asks them, because their role is defined as "build what we tell you to build," not "tell us what we should build." The cost of excluding MOPs from strategy When marketing ops is excluded from strategic decisions, those decisions get made without the one team that understands the operational reality. The CMO decides to launch an ABM programme. Nobody asks MOPs whether the data architecture supports account-level targeting. It doesn't. The team spends three months building workarounds. Leadership decides to migrate platforms. Nobody asks MOPs about the complexity of the current automation environment. The migration timeline is set at three months. It takes nine. The extra six months weren't caused by the new platform being difficult - they were caused by the old platform being far more complex than anyone outside MOPs understood. Someone in the leadership team approves a new tool. Nobody asks MOPs whether it integrates with the existing stack. It doesn't — not cleanly. The team spends weeks building a custom integration that a five-minute conversation would have flagged before the purchase. These aren't hypothetical scenarios. They're patterns that repeat in every organization that treats MOPs as execution rather than strategy. The decisions get made upstairs, the consequences get absorbed downstairs, and the team that could have prevented the problem wasn't in the room when the decision was made. What MOPs actually knows Marketing ops sits at the intersection of marketing strategy, data, technology, and revenue operations. No other function in the organization has that cross-functional visibility. MOPs knows which campaigns are actually driving pipeline - not what the dashboard says, but what the data actually shows when you dig past the vanity metrics. They know which segments are engaged and which are exhausted. They know which parts of the lead lifecycle are working and which are leaking. They know where the data is clean and where it's not. They know which integrations are stable and which are held together with workarounds. They know which automations are running as intended and which have drifted. They know where the compliance risks sit - which consent records are current, which suppression rules make sense, and which AI features are running without anyone monitoring them. This knowledge isn't just operational - it's strategic. A CMO who understands the state of their marketing infrastructure, data quality, and automation environment makes better decisions than one who doesn't. And the only team that can provide that understanding is MOPs. But when MOPs is buried in the org chart as a service desk, that knowledge never reaches the people making decisions. It stays trapped in the team that has it, used only to react to problems instead of prevent them. The org chart problem Where MOPs sits in the organization determines what it's allowed to do. And in most companies, MOPs sits too low. When MOPs reports to a demand gen manager, the team's priorities get set by campaign timelines and lead targets. There's no mandate to audit, optimize, or advise - just to build and send. The work is defined by the queue, and the queue is defined by someone whose job is to generate leads, not to build operational infrastructure. When MOPs reports to a VP or director of marketing operations - or better, to the CMO directly - the team's mandate expands. They can prioritize platform health alongside campaign execution. They can flag data quality issues before they become pipeline problems. They can advise on technology decisions before the purchase, not after. They can build the governance, documentation, and process infrastructure that every organization needs and nobody wants to fund. The reporting line doesn't just affect the team's authority. It affects what the organization sees as MOPs' purpose. If MOPs reports to a campaign manager, MOPs is a campaign support team. If MOPs reports to the CMO, MOPs is an operational function with strategic input. The team's capabilities don't change. The organization's willingness to use them does. What changes when MOPs gets a strategic seat The shift isn't dramatic. It doesn't require a reorganisation or a new title. It requires including MOPs in the conversations where decisions get made - and then actually listening to what they say. Before a platform purchase, MOPs evaluates the integration requirements and flags complications the vendor won't mention. Before a migration, MOPs maps the current automation environment so the timeline reflects reality. Before an ABM launch, MOPs assesses whether the data supports account-level targeting. Before an AI feature gets activated, MOPs checks what data it consumes and whether that data is reliable. These are five-minute conversations that save months of rework. But they only happen when MOPs is in the room - and when the organization recognizes that the team building the campaigns also understands the infrastructure those campaigns depend on. The best marketing operations teams aren't the ones that build the fastest. They're the ones that get asked "should we do this?" before they get told "build this." That question is the difference between a support function and a strategic one. The team you're underusing is the one that knows the most Every organization that's invested in a marketing automation platform, a CRM integration, a data infrastructure, and a campaign operation has already invested in marketing ops - whether they think of it that way or not. The team exists. The knowledge exists. The cross-functional visibility exists. The question is whether the organization uses that investment fully or wastes it by limiting MOPs to building emails and pulling reports. One path produces a marketing operation that's reliable, scalable, and strategically informed. The other produces a service desk that's permanently overwhelmed and permanently undervalued. The platform doesn't care who decides the strategy. But the strategy works better when the people who understand the platform are involved in making it. That's not a radical idea. It's just one that most organizations haven't acted on yet.

The EU AI Act becomes broadly applicable on 2 August 2026. For some organisations, that date will represent the final stage of a carefully managed programme involving legal, security, IT, data and every business team using AI. For others, it will mark the beginning of a frantic search for a spreadsheet somebody vaguely remembers creating last year. Your Marketing Operations should probably start checking which camp it is in. Because while the conversation about AI compliance has largely been happening in boardrooms, legal teams and technology departments, much of the actual use is happening inside marketing. It is embedded in platforms. Added to campaign processes. Connected to customer data. Used to create content, prioritise accounts, recommend actions, personalise experiences and communicate directly with prospects. In some cases, it is doing all of that without anybody having formally decided that it should. That is the uncomfortable part. The biggest immediate challenge for Marketing Operations may not be understanding every article of the EU AI Act. It may be working out what the organisation is already using, what those systems are doing and who is responsible for them. Marketing did not wait for the governance meeting AI did not enter most marketing organisations through a carefully controlled transformation programme. It arrived through product updates. A copywriting feature appeared inside a campaign platform. A meeting tool started producing summaries. A media platform introduced automated creative. A salesperson connected a browser extension to the CRM. Somebody uploaded a customer list into a tool to “see what it could do.” Then came the pilots, custom assistants, automated workflows and agents. Each individual decision may have seemed small. Together, they have created a network of systems using company information, customer data and business rules in ways that are not always visible from the centre. The problem is not necessarily that all of this activity is reckless. Some of it may be entirely sensible and low risk. The problem is that many organisations cannot describe it accurately. Ask which AI systems marketing uses and you may receive a list of officially purchased tools. That is not the same thing. The real list also includes AI features inside existing platforms, free tools used by individuals, systems trialled by agencies, functions switched on by vendors, integrations created by employees and automations nobody has looked at since the person who built them left. If your inventory only includes products with “AI” in the contract title, it is probably already wrong. Buying the platform does not settle ownership One of the easiest mistakes is assuming the technology team owns anything involving AI. It may own the contract. It may manage access. It may review the security. None of that means it understands how marketing is using the system. Legal may interpret the regulation, but legal does not build campaign workflows. IT may approve the platform, but IT does not decide which customer data should be used for personalisation. Procurement may negotiate the agreement, but procurement does not know whether an automated recommendation is being treated as an interesting suggestion or as an instruction that nobody questions. The team using the system owns part of the responsibility because it owns the business context. In Marketing Operations, that context matters. A tool generating rough ideas for internal campaign planning is not doing the same job as a system deciding which people receive an offer. A feature correcting grammar is not the same as a chatbot communicating directly with customers. A model suggesting possible target accounts is not the same as a process automatically excluding people from an opportunity. The technology may look similar on a systems diagram. The consequences are not. That means Marketing Operations cannot simply hand the entire subject to legal and wait for a policy document. It needs to explain what the systems actually do. You cannot govern an invisible stack The first practical job is not writing a 70-page AI policy. It is finding the technology. Marketing Operations needs an inventory that reflects reality rather than the approved software catalogue. For each system, the organisation needs to know what it does, which team uses it, what information it can access, what it produces and whether its output affects customers, employees or business decisions. It also needs a named owner. “Marketing” is not an owner. “The automation team” is not much better. Ownership needs to reach an identifiable person who understands the use case and can answer questions about it. That does not mean this person carries every legal obligation alone. It means somebody is responsible for making sure the system does not disappear into the organisational wallpaper. The inventory should also cover features inside platforms the organisation already owns. This is where things get messy. Software providers are racing to add AI functions to almost everything, often enabled through ordinary product releases. A platform that was reviewed two years ago may now behave quite differently. The contract may not have changed. The risk may have. Marketing Operations should therefore be asking vendors direct questions. Which features use AI? What data do they access? Is customer information used to improve external models? Can the feature be disabled? Are actions logged? Can a human review the output? What happens when the system gets something wrong? A shiny product page containing the words “responsible” and “enterprise-grade” is not an adequate answer. The deadline is not the starting gun There is also a dangerous assumption that organisations have until August to begin thinking about this. They do not. Some requirements are already applicable, including the obligation for organisations providing or using AI systems to ensure that relevant staff have a sufficient level of AI literacy. That does not mean every marketer needs to become a machine-learning engineer. It means people should understand enough about the systems they use to recognise their limitations, apply appropriate judgement and avoid creating obvious harm. A generic one-hour training course followed by a multiple-choice quiz may produce a completion certificate. It does not necessarily produce competent use. The training should reflect the job. A content writer needs to understand accuracy, attribution, confidentiality and the risks of publishing generated material without proper review. A campaign manager needs to understand what can happen when a system creates segments, selects audiences or changes workflows. A Marketing Operations leader needs to understand permissions, data access, monitoring, approval processes and accountability. The person connecting a tool to the CRM needs considerably more than a reminder not to paste passwords into a chat window. Training should match what people are actually allowed to do, otherwise, the organisation has technically educated everybody while practically preparing nobody. Transparency is about more than adding a disclaimer From 2 August 2026, transparency obligations under the EU AI Act will apply to certain systems and content. For marketing teams, this is likely to bring particular attention to customer-facing chatbots, synthetic images, video or audio, deepfake-style material and some AI-generated text relating to matters of public interest. This does not mean every AI-assisted email subject line needs a warning label large enough to frighten the recipient. It does mean organisations need to understand where disclosure is required and ensure the process exists to make it happen. That process cannot rely entirely on the person publishing the content remembering to tick a box. Marketing Operations should help build disclosure and review requirements into workflows. Where content must be identified, the system should support it. Where a customer is interacting with a machine, that should not be hidden behind deliberately vague language and a stock photograph of someone wearing a headset. Transparency is not just a legal inconvenience. It is part of preserving trust. Most customers will accept that organisations use automation. What they will not appreciate is feeling tricked. The hidden issue is decision-making Content generation receives most of the attention because it is visible. The harder questions sit underneath it. What is the organisation allowing AI to decide? Marketing systems increasingly recommend audiences, prioritise accounts, predict behaviour, adapt journeys, score leads and select the next action. Again, not every automated marketing decision falls into the most heavily regulated category. Claims that every lead score is suddenly a major EU AI Act emergency are not particularly helpful. But the absence of a dramatic legal classification does not make a process sensible. Marketing Operations should still understand which decisions are automated, which are influenced by automated recommendations and where human judgement remains. It should be possible to explain why a person entered a particular journey, received a specific message or was excluded from an opportunity. It should also be possible to challenge the system. If the team treats every recommendation as correct because it arrived inside a polished dashboard, there is no meaningful oversight. There is just a human clicking “approve” to make the workflow look respectable. Governance should improve the work This is the point where many organisations make governance unnecessarily painful. They create committees, forms, approval stages and policy documents without fixing the way work happens. Employees then find unofficial routes around the process because the official one takes six weeks and requires a meeting with fourteen people. Good governance should make acceptable uses easier and questionable uses harder. People should know which tools are approved, what they can use them for and when additional review is required. There should be a straightforward route for proposing a new use case. Higher-risk activity should receive closer scrutiny. Routine, low-impact activity should not require an emergency summit. Marketing Operations is well placed to help design this because it already understands workflows, permissions, quality assurance and operational controls. Or at least it should. If the team can build a 47-step nurture programme with six branches and three regional exceptions, it can probably design a sensible approval process for an AI-enabled campaign. What Marketing Operations needs to do now The immediate priority is visibility. Find the systems. Document the use cases. Identify the data involved. Assign owners. Review access. Check vendor terms. Understand which outputs reach customers and which decisions are being influenced. Then look at the controls. Can people review outputs before they are used? Are important actions logged? Is there a clear escalation route? Can the organisation stop the system quickly? Does somebody periodically check whether it is still doing what it was intended to do? Finally, look at the people. Do they understand the tools they use? Do they know what information they can enter? Do they know when human review is mandatory? Would they recognise a poor or inappropriate output, or are they trusting the system because it sounds confident? None of this requires Marketing Operations to become the legal department. It requires the team to behave like the operational owner of marketing technology. Which, inconveniently, is exactly what the name suggests. August will expose the gaps, not create them The EU AI Act is not suddenly going to make poorly controlled technology risky on 2 August. That risk already exists. The deadline simply makes it harder for organisations to continue pretending that nobody owns it. Marketing Operations has an opportunity here. It can wait for legal to send around a policy and then attempt to bolt it onto a marketing stack that nobody has fully mapped. Or it can take the lead in understanding how AI is actually being used, where it touches customers and data, and what practical controls need to exist. That is not bureaucratic housekeeping. It is the difference between using AI as part of a functioning operation and scattering it across the business until something embarrassing forces everyone to pay attention. The deadline is approaching... The first question is not whether your organisation is compliant. It is whether Marketing Operations even knows what it owns. Discover our AI Services

There's a particular kind of panic that sets in when a competitor launches something impressive. A slick new campaign. A personalized experience that feels like it was built by a team twice your size. A webinar series that runs weekly without missing a beat. Content that shows up everywhere, perfectly timed, perfectly targeted. The instinct is to assume they have better tools. A more powerful platform. A bigger tech budget. Some integration or capability you don't have access to. So the conversation starts: what are they using? Should we switch platforms? Do we need to buy something new? Almost always, the answer is no. They're not using better technology. They're using the same technology - or something very similar - with better processes underneath it. The platform is the same. The difference is how they run it. Same tools, different outcomes Most B2B marketing teams in any given industry are running variations of the same stack. The same handful of MAPs. The same CRM. Similar enrichment tools, analytics platforms, and advertising channels. The technology landscape has consolidated enough that the tools available to a mid-market company are functionally similar to the tools available to an enterprise. The differentiation isn't in the tools. It's in how they're configured, maintained, and operated. One team builds campaigns in two hours because they have a brief template that arrives complete, a template system that works reliably, an approval process with defined turnaround times, and a QA checklist that catches problems before send. Another team builds the same campaign in two weeks because the brief arrives vague, the templates are broken, approvals get stuck in email chains, and QA is someone squinting at a test send on their phone. Both teams are using the same platform. The output gap has nothing to do with technology. It's entirely about the process surrounding it. Process isn't exciting. That's why it works. Nobody gets promoted for building a campaign brief template. Nobody presents a QA checklist at the company all-hands. Nobody writes a LinkedIn post about how they redesigned the approval workflow. Process work is invisible when it works and only noticed when it's absent. Which is exactly why most teams don't prioritize it. The team is rewarded for campaigns launched, pipeline generated, content produced - visible outputs that show up in reports and reviews. The operational infrastructure that makes those outputs possible doesn't get measured, doesn't get celebrated, and doesn't get resourced. This creates a cycle where the team is always busy but never efficient. Every campaign takes longer than it should because the process hasn't been fixed. But nobody fixes the process because the team is too busy building campaigns. The urgent always wins over the important, and the process debt compounds quarter after quarter. Meanwhile, the competitor who invested a week in fixing their brief template, another week in rebuilding their email templates, and another week in structuring their approval workflow is now building campaigns in a fraction of the time. They didn't buy anything new. They fixed the machine they already had. Where process failures actually live The process problems that cost the most time aren't dramatic. They're mundane. They're so routine that the team has stopped seeing them as problems and started accepting them as "just how things work." The brief that arrives incomplete. Every missing field on a campaign brief turns into a conversation - a Slack message, an email, a meeting to clarify what should have been specified upfront. Multiply that by every campaign and the wasted hours are staggering. A brief template with required fields and a "this goes back if it's incomplete" rule eliminates most of it. The template that requires workarounds. If the person building the campaign spends 30 minutes per build working around template limitations, that's 30 minutes multiplied by every campaign for the rest of the year. Rebuilding the templates is a one-time cost that pays back permanently. The approval chain with no deadlines. An approval workflow where the expected turnaround is "whenever they get to it" is an approval workflow that adds days to every build. Setting a 24-hour review window for each stage - with escalation if it's missed - compresses weeks into days. The QA process that depends on memory. If QA is "the builder checks everything they can think of," different builders will check different things and something will eventually get missed. A shared QA checklist takes an hour to build and ensures consistent quality on every send. The handoff between teams that nobody designed. Marketing produces the MQL. Sales receives it. But how? Through what mechanism? With what context? On what timeline? If the handoff isn't defined - if it's just a notification in the CRM that sales may or may not see - the entire upstream process loses value at the exact point where it should be generating it. Each of these is a small problem. Combined, they're the reason one team operates at twice the speed of another using the same tools. The process audit most teams skip Technology audits are common. Someone reviews the stack annually, evaluates new tools, and makes recommendations. Process audits almost never happen - which is strange, because process problems cost more time than technology problems in most marketing operations. A process audit looks at how work actually flows through the team, not how it's supposed to flow. Map the journey of a campaign from request to send. How many handoffs are there? How many of those handoffs involve waiting? Where do things get stuck? What are the most common reasons a campaign gets delayed? Then do the same for lead management. When a lead becomes an MQL, what happens? How long does it take to reach sales? What context arrives with it? What percentage get followed up within 24 hours, 48 hours, a week? Where does the process break down? The answers are almost always embarrassing. Not because the team is incompetent - because the process was never designed. It evolved through individual decisions made under time pressure, and nobody stepped back to look at the whole picture. Technology is a ceiling. Process is the floor. Your marketing automation platform defines what's possible. Your process defines what actually happens. Most teams are operating well below what their platform can do - not because they need more features, but because the operational infrastructure around the platform hasn't been built to the same standard as the technology. The competitor who looks like they're running a better operation probably is. But they're not running better technology. They're running better processes - clearer briefs, faster approvals, reliable templates, consistent QA, designed handoffs. None of that cost them a new licence fee. All of it cost them a few weeks of operational work that nobody wanted to do but everyone benefits from. The gap between your team and the one you're envying isn't in the tools. It's in the operational discipline underneath them. And unlike technology, process improvements don't require a budget approval, a vendor evaluation, or a six-month implementation. They require someone deciding that how the team works matters as much as what the team produces. That decision is usually the hardest part. Everything after it is straightforward.

Somewhere in your marketing technology stack, there's a tool that nobody uses. Not "underused." Not "we use it for one thing." Nobody uses it. The login credentials have been lost. The integration was never completed. The person who championed the purchase left the company eight months ago. The annual renewal went through automatically because finance didn't flag it and nobody in marketing remembered to cancel it. That tool is the most expensive thing in your stack - not because of the licence fee, but because of what it represents. A problem that was identified, a purchase that was approved, an implementation that never happened, and a decision nobody wants to revisit because cancelling it means admitting the investment was a mistake. Every stack has at least one This isn't an edge case. Talk to anyone who's run a martech audit on an enterprise marketing department and they'll tell you the same thing: every stack has tools that nobody can justify. Not one or two - usually several. The pattern is always the same. Someone identified a gap. A vendor appeared with a compelling demo. The purchase got approved based on a use case that sounded reasonable at the time. Then reality set in. The implementation was harder than expected. The integration with existing tools required development resources that weren't available. The team that was supposed to use it was already stretched thin. The vendor's customer success team checked in for the first month and then disappeared. So the tool sat there. Not producing value. Not integrated. Not cancelled. Just existing - a line item on a budget that nobody reviews closely enough to question. The real cost isn't the licence The licence fee is the obvious cost, but it's rarely the biggest one. The bigger cost is the problem that's still unsolved. The tool was purchased to fix something - a gap in reporting, a missing integration, a capability the team needed. When the tool went unused, the gap didn't close. It just stopped being talked about. The team found workarounds, or they accepted the limitation, or they forgot the gap existed because the purchase had created the illusion of progress. Then there's the opportunity cost. Every pound spent renewing a tool nobody uses is a pound not spent on something that would actually help - better configuration of existing platforms, training for the team, consulting to fix the processes that are actually broken. And there's the complexity cost. Even unused tools add to the stack's overhead. They show up in security audits. They hold data that may need to be accounted for under privacy regulations. They create confusion when a new team member sees them in the tool inventory and asks what they're for - and nobody can answer. Why nobody cancels The psychology of unused tools is straightforward: cancelling is harder than renewing. Cancelling requires someone to acknowledge that the purchase was a mistake - or at minimum, that the implementation didn't work. That's uncomfortable. The person who approved the budget might still be in the room. The vendor relationship might feel awkward to end. The original use case might still technically be valid, even if nobody's going to act on it. Renewing requires nothing. The payment processes automatically. Nobody has to make a decision, have a conversation, or write an email. Inaction is the path of least resistance, and in most organisations, the systems are set up to make inaction the default. This is how tools stay in the stack for years after they stopped being relevant. Not because anyone decided to keep them, but because nobody decided to remove them. One unused tool becomes three The problem compounds. An unused tool doesn't just sit there costing money - it creates conditions for more unused tools to arrive. When the original tool fails to deliver, the gap it was supposed to close remains open. The team still has the problem. So they start looking for another solution - a different tool, a different vendor, a different approach to the same issue. Sometimes that new tool works. Often it joins the first one in the stack, partially implemented, partially adopted, partially solving the problem. Meanwhile, the platform the team already owns - the MAP, the CRM, the analytics tool - may have added the exact capability the unused tool was supposed to provide. Platform vendors ship new features constantly. The enrichment tool purchased two years ago might now be redundant because the MAP added native enrichment. The reporting dashboard bought last year might be unnecessary because the CRM's built-in analytics improved significantly. But nobody checks, because nobody's tracking the overlap between new platform features and existing point solutions. After a few cycles of this, the stack has multiple tools doing overlapping things, none of them fully integrated, and the team is spending more time managing tool complexity than solving the problem any of them were supposed to address. The vendor renewal conversation Most tool renewals happen without a conversation. An invoice arrives, finance processes it, and the subscription continues. The vendor doesn't check whether you're using the tool - they're happy to keep billing. Your team doesn't flag it because nobody is tracking utilisation against cost. Building one step into the renewal process changes this entirely. Thirty days before any tool renews, someone - the tool owner, the ops lead, whoever manages the stack - should answer four questions and document the answers. Is the tool actively used? Not "does someone log in occasionally" - is it part of a regular workflow that would break if the tool disappeared? If the last meaningful usage was three months ago, that's not active. Is the tool integrated with the rest of the stack? A standalone tool that doesn't connect to the CRM, the MAP, or the reporting infrastructure is a tool operating in isolation. Isolated tools produce isolated data. That's rarely worth the licence fee. Has the original use case been addressed by another tool since purchase? Platform capabilities evolve. Check whether the MAP, the CRM, or another tool in the stack has added functionality that makes this tool redundant. Does the cost justify the value? Not in theory - in practice. What specific business outcome did this tool contribute to in the last 12 months? If the answer requires creative interpretation, the tool probably isn't worth the renewal. The audit nobody wants to do The fix is simple and uncomfortable: inventory every tool in your stack, identify who uses each one, and cancel what can't be justified. For each tool, ask three questions. Who on the team used this in the last 90 days? If nobody, that's your answer. What business outcome does this tool directly support? If the answer is vague or theoretical, the tool isn't supporting anything. If we cancelled this tomorrow, what would break? If the answer is "nothing," cancel it tomorrow. Most teams that run this exercise for the first time are surprised by how much they're spending on tools that deliver no measurable value. The savings are usually significant - not life-changing, but meaningful enough that the budget could be redirected somewhere useful. The harder part is building the habit. Tool audits should happen at least annually, timed to renewal cycles. Before any tool renews, someone should confirm it's still being used, still integrated, and still solving the problem it was purchased to solve. If the answer to any of those is no, the renewal should be a conscious decision - not an automatic one. The stack should shrink before it grows Every time someone proposes adding a new tool, the first question should be: have we fully used what we already have? The second question should be: is there anything in the current stack that could be removed to make room? A lean stack that's properly configured, fully integrated, and actively used by the team will outperform a bloated one every time. The tools that create the most value aren't the newest or most feature-rich. They're the ones that someone actually owns, maintains, and uses every day. The most expensive tool in your stack isn't the one with the biggest licence fee. It's the one nobody would miss if it disappeared - and the fact that it's still there tells you more about your procurement process than it does about your marketing technology strategy.

The quarterly marketing review follows the same script in most organizations. Someone opens a slide deck. The first few slides show activity metrics - emails sent, campaigns launched, webinars hosted, content pieces published. Then engagement metrics - open rates, click rates, form submissions, social impressions. Then the MQL number, presented with a slight uptick and a green arrow. Leadership nods. Someone asks a question about pipeline. The answer involves the word "influenced" used loosely. The meeting ends. Everyone goes back to their desk knowing that nothing in that presentation answered the question leadership was actually asking: is marketing generating revenue? The metrics weren't wrong. They were real numbers from real campaigns. But they were the wrong numbers - activity and engagement metrics dressed up as performance metrics, presented with enough confidence that nobody in the room wanted to be the one to say "but what did any of this actually produce?" The gap between what gets reported and what gets asked Leadership asks one question about marketing: is it working? Specifically, is the money we're spending on marketing producing a return? Is the pipeline growing? Are deals closing faster? Is marketing contributing to revenue in a way that justifies the investment? The metrics most marketing teams report don't answer that question. They answer a different one: is the marketing team busy? Emails sent tells you the team is active. It doesn't tell you whether those emails produced anything. Open rates tell you subject lines are working. They don't tell you whether anyone who opened the email went on to become a customer. MQLs tell you leads are crossing a threshold. They don't tell you whether those leads converted to opportunities, entered the pipeline, or generated a single pound of revenue. These metrics aren't useless. They're operational indicators - useful for the marketing team to diagnose and optimize their own campaigns. But they're not performance metrics. They don't connect to the business outcomes that leadership cares about. Presenting them as if they do is where the credibility gap starts. Why teams report this way It's not because marketers are dishonest. It's because revenue-connected reporting is hard to build and most teams don't have the infrastructure to do it properly. Connecting marketing activity to pipeline and revenue requires clean data flowing between the MAP and the CRM. It requires attribution models that are configured correctly and maintained over time. It requires lifecycle stages that are consistently defined and applied. It requires closed-loop reporting that tracks a lead from first touch through to closed-won deal. Most teams don't have all of that in place. The CRM integration has gaps. The attribution model was set up once and never reviewed. Lifecycle stages mean different things to different people. The data that would connect email click to pipeline contribution doesn't exist because the handoff between marketing and sales isn't tracked cleanly. So the team reports what they can measure - activity and engagement - because those numbers are available, they're always positive (you can always send more emails), and they fill a slide deck without requiring anyone to confront the harder question of whether any of it drove revenue. Over time, this becomes the norm. Leadership stops expecting revenue metrics from marketing because they never get them. Marketing stops trying to build them because leadership stopped asking. Both sides settle into a comfortable arrangement where marketing reports activity, leadership acknowledges it, and the actual impact question goes unasked and unanswered. The damage is slow and structural Vanity metrics don't cause an immediate crisis. They cause a slow erosion of marketing's credibility and strategic influence within the organization. When marketing can't connect its activity to revenue, it gets treated as a cost centre - a department that spends money rather than one that generates it. Budget conversations become adversarial. Every investment requires justification, and the justification can never be "this will generate pipeline" because marketing can't prove the last investment did. When times get tight and cuts need to happen, the departments that can't demonstrate revenue contribution get cut first. Marketing teams that report vanity metrics are perpetually vulnerable to budget reductions because they've never built the evidence base that protects them. And the strategic seat at the table disappears. Leadership doesn't invite marketing into revenue conversations because marketing has never shown it belongs there. The CMO gets left out of planning discussions, pipeline reviews, and forecasting meetings - not out of malice, but because marketing's reporting has never demonstrated a connection to the numbers being discussed in those rooms. What revenue-connected reporting actually looks like The shift from vanity metrics to revenue metrics isn't a technology problem. It's a definition problem followed by a configuration problem. Start by defining what you're going to measure. The metrics that connect marketing to revenue are straightforward: marketing-sourced pipeline (deals where the first meaningful touch came from marketing), marketing-influenced pipeline (deals where marketing touched one or more contacts during the sales cycle), MQL-to-opportunity conversion rate, and average deal velocity for marketing-sourced vs non-marketing-sourced deals. None of these require exotic tools. They require clean data, a properly configured CRM integration, and agreement between marketing and sales on what counts as "sourced" and "influenced." Build the attribution model and maintain it. First-touch, last-touch, multi-touch - the specific model matters less than having one that's configured correctly and reviewed regularly. The model should reflect your actual buyer journey, not a theoretical one. If most of your deals involve 8-10 marketing touches before the first sales conversation, a first-touch model is going to undercount marketing's contribution. If your sales cycle is short and driven by a single conversion event, multi-touch may be overcomplicating things. Close the loop between the MAP and the CRM. When a lead becomes an opportunity and when that opportunity closes, the data should flow back to marketing so the original campaign, channel, and touchpoints get credited. Without this closed loop, marketing can report on what it sent but never on what it produced. Report at the level leadership cares about. The QBR slide deck should lead with pipeline and revenue metrics. How much pipeline did marketing source this quarter? How much did it influence? What's the conversion rate from MQL to opportunity? How does deal velocity compare for marketing-sourced vs other deals? Activity and engagement metrics can follow as supporting detail - but they're the footnotes, not the headline. The conversation nobody wants to have The first time you present revenue-connected metrics to leadership, the numbers might not look great. Marketing-sourced pipeline might be smaller than expected. Conversion rates might reveal that the MQL definition needs work. Attribution might show that some of the campaigns leadership loves aren't actually producing results. That's uncomfortable. It's also the beginning of marketing being taken seriously as a revenue function. The teams that make this shift - from reporting what they did to reporting what it produced - earn a fundamentally different relationship with leadership. Budget conversations become investment conversations. Marketing gets pulled into pipeline reviews. The CMO gets a seat in the rooms where revenue is discussed. The teams that don't make the shift keep presenting slide decks full of open rates and MQL counts. Leadership keeps nodding. And everyone keeps leaving the room knowing the actual question wasn't answered. The metrics are available. The infrastructure is buildable. The only thing stopping most teams is the willingness to report honestly - even when the honest numbers are harder to celebrate than the vanity ones.

Here's how most ABM programmes actually work: someone builds a target account list in a spreadsheet. The list gets uploaded into the marketing automation platform or the ABM tool. The same campaigns that were already running - the same nurture emails, the same content offers, the same webinar invitations - get filtered to only hit contacts at those accounts. The team calls it ABM. Leadership reports it as ABM. The ABM platform vendor counts it as adoption. But nothing about the approach actually changed. It's the same demand generation programme with a filter applied. The strategy is identical. The content is identical. The measurement is identical. Only the audience narrowed. That's not account-based marketing. That's demand gen wearing a name badge. What ABM is supposed to be The entire premise of account-based marketing is that you treat individual accounts as markets of one. You research the account. You understand their specific challenges, their organizational structure, their buying committee. You build campaigns tailored to that account's reality - not your generic messaging repackaged with their logo on it. That means different content for different stakeholders within the same account. The CFO gets messaging about financial impact. The IT director gets messaging about integration and security. The end-user team gets messaging about workflow improvements. Each stakeholder receives something relevant to their role in the buying decision, timed to where the account is in its evaluation process. It also means sales and marketing working the account together - not marketing generating leads and throwing them over the wall. In real ABM, the sales team and marketing team agree on the account plan, coordinate outreach, share intelligence about what's happening inside the account, and adjust the approach based on what they're learning in real time. That level of coordination is hard. It takes planning, resources, content, and genuine collaboration between teams that in most organizations operate independently. Which is why most teams skip it and just filter their existing campaigns by an account list. The spreadsheet is not a strategy The target account list is where ABM starts. It's not the strategy itself. But in most programmes, building the list is the only genuinely account-based activity that happens. Everything after it is generic. A good target account list is built on data - closed-won analysis, firmographic fit, intent signals, strategic value, sales input. That part usually gets done reasonably well because it's a finite, one-time exercise that produces a deliverable everyone can point to. The strategy is what happens after the list exists. How are you engaging each tier of accounts differently? What content exists for each persona in the buying committee? How does marketing activity coordinate with sales outreach? What signals indicate an account is progressing, and what actions do those signals trigger? How are you measuring engagement at the account level, not just the lead level? If the answers to those questions are vague - or identical to how you'd answer them for your demand generation programme - the ABM label isn't earned. The content problem Real ABM requires content that most marketing teams don't have and aren't set up to produce. Demand gen content is built for scale - one ebook serves the entire addressable market, one webinar targets a broad audience, one email template gets sent to thousands of contacts with light personalization. That's efficient and it works for demand gen. It doesn't work for ABM. ABM content needs to be relevant at the account level or at minimum the industry and persona level. That means a case study that speaks to the specific challenges of financial services companies, not a generic customer story. A whitepaper that addresses the regulatory environment the target account operates in, not a broad trends piece. An email that references something specific about the account's situation, not a merge field with their company name dropped in. Producing this content takes significantly more effort per account than producing demand gen content per segment. Most teams underestimate this when they launch ABM. They commit to the strategy, build the account list, and then discover they don't have the content to support account-specific engagement. So they fall back on the generic content they already have - and the programme becomes demand gen with a filter again. Sales alignment isn't optional - it's the whole point The most common structural failure in ABM programmes isn't bad targeting or weak content. It's that sales isn't involved. Marketing builds the account list. Marketing runs the campaigns. Marketing tracks the engagement scores. Sales gets a notification that an account is "engaged" and does whatever they were going to do anyway - which is usually calling the one contact they already know and ignoring the rest of the buying committee. That's not alignment. That's parallel play. Marketing and sales are both active on the same accounts, but they're not coordinating. The messaging isn't consistent. The timing isn't coordinated. The intelligence isn't shared. Marketing doesn't know what sales is hearing in conversations. Sales doesn't know which stakeholders marketing has engaged. ABM without sales alignment is marketing talking to itself about accounts. The investment in targeting, content, and technology gets wasted because the last mile - the human relationship between the sales team and the buying committee - never connects to the marketing activity that's supposed to support it. The measurement theatre Measurement is where the illusion gets maintained. Most ABM programmes report metrics that sound account-based but are actually demand gen metrics with a filter. "We generated 150 MQLs from target accounts this quarter." That's a demand gen metric applied to an account list. It doesn't tell you anything about account penetration, buying committee coverage, or whether the accounts are actually progressing toward a deal. "Our target account engagement score increased by 30%." Engagement scoring at the account level is a step in the right direction - but what does the score actually measure? If it's aggregating email opens and content downloads, it's measuring marketing activity, not buying intent. An account where one person downloaded three ebooks isn't more engaged than an account where five decision-makers each visited the pricing page once. But most engagement models would score the first account higher. "We influenced pipeline worth £2M from ABM accounts." Influenced is doing a lot of heavy lifting in that sentence. Was the account already in pipeline before ABM started? Would sales have closed it anyway? Did the ABM activity actually change anything about the deal, or did it just happen to touch an account that was already progressing? Real ABM measurement is harder and more honest. It tracks how many stakeholders in the buying committee have been engaged, whether engagement is progressing across the account over time, whether ABM-targeted accounts enter pipeline at a higher rate than non-targeted accounts, and whether they close faster or at higher values. If your ABM reporting can't answer those questions, it's reporting on demand gen and calling it ABM. What the first 90 days of real ABM look like If your honest self-assessment revealed that your programme is closer to demand gen with a filter, here's what actually shifting to ABM looks like in practice. Not the full transformation - just the first 90 days. Days 1-30: Shrink the list and deepen the research. Take your target account list and cut it by at least half. The accounts that remain should be ones you can genuinely research and build tailored approaches for. For each one, map the buying committee - not just the contact you already have, but the full set of stakeholders who would be involved in a purchase decision. Use your ABM platform, LinkedIn, and your sales team's relationships to build that map. Days 30-60: Build account-specific content for one tier. Pick your top 10-20 accounts and build content that speaks to their specific industry, challenges, or situation. This doesn't mean creating a custom ebook for each account. It means adapting your best existing content to address the specific concerns of each buying committee persona within that industry. The CFO version. The IT version. The practitioner version. Three versions of one asset is more valuable than one generic version sent to everyone. Days 60-90: Coordinate one joint campaign with sales. Pick five accounts and run a coordinated play where marketing and sales are actively collaborating. Marketing warms the account with targeted content and ads. Sales follows up with personalized outreach that references the same themes. Both teams share what they're seeing - which stakeholders are engaging, what topics are resonating, where the gaps are. Run this for 30 days and measure what happens compared to your standard approach. That's not a complete ABM programme. It's a proof of concept that demonstrates whether genuine account-based activity produces different results from filtered demand gen. If it does - and it usually does - you have the evidence to invest further. If it doesn't, either the execution needs adjusting or the accounts weren't the right ones. How to know if your ABM is actually ABM Honest self-assessment. Four questions. Can you describe a different approach for your top 10 accounts vs your top 100? If everything gets the same treatment, you're running one-to-many demand gen, not tiered ABM. Does sales co-own the account plan, or just receive the leads? If sales isn't involved in deciding which accounts to target, what messaging to use, and how to coordinate outreach, ABM is a marketing-only initiative - and marketing-only ABM doesn't close deals. Do you have content tailored to specific industries, personas, or accounts? If every account receives the same content, the personalization is cosmetic. Real ABM requires real content investment. Are you measuring account engagement, or just lead engagement? If your primary metric is MQLs from target accounts, you're still measuring demand gen. ABM measures how deeply you've penetrated the buying committee, how engagement is progressing at the account level, and how that engagement connects to pipeline. If you answered honestly and the answers were uncomfortable, you have two choices. Either invest in building a real ABM programme - with the content, the sales coordination, and the measurement to match - or acknowledge that what you're running is demand gen with better targeting and stop calling it ABM. Both are valid strategies. Only one of them is account-based marketing.

There's plenty of content about why AI governance matters. The regulatory pressure, the risk of ungoverned automation, the compliance deadlines. We've written about it ourselves. But there's far less content about what good AI governance actually looks like in practice - inside a real marketing operations environment, with a real team, running real campaigns on a real platform. This is that article. Not the principles. Not the framework diagram. The operational reality of what it looks like when a marketing ops team is actually governing their AI properly - and how it's different from what most teams are doing. What most teams are doing Most B2B marketing teams have some version of the following in place: an AI policy document, a general awareness that AI features exist in their platform, and a vague understanding that someone should probably be paying attention to what those features are doing. In practice, AI features get activated and forgotten. Nobody maintains a list of what's running. Nobody reviews outputs. Nobody checks whether the data feeding AI features is still current. Nobody owns the AI layer as a distinct operational responsibility. The team treats AI features the same way they treat any other platform capability - configure it, trust it, move on. This works until it doesn't. When it stops working, the failures are hard to detect and harder to diagnose, because nobody built the infrastructure to monitor what the AI is doing. What a well-governed team actually does differently The difference isn't dramatic. There are no dedicated AI governance departments or six-figure compliance platforms. The difference is a small set of operational habits that take maybe 2-3 hours per month and produce a level of visibility and control that most teams don't have. They maintain a live inventory. Every AI feature running inside the platform is listed in a shared document - not a one-time audit, but a living register that gets updated whenever something changes. The register is simple: feature name, what it does, what data it consumes, when it was activated, who owns it, when it was last reviewed. When a platform upgrade introduces new AI capabilities, someone checks what changed and updates the register. When a team member activates a new feature, they add it. When a feature gets deactivated, it's noted. This takes minutes per update and creates something invaluable: a single source of truth for what AI is doing inside the platform. Most teams can't produce this list even after hours of investigation. A well-governed team can produce it in 30 seconds. They assign owners, not committees. Each AI feature in the register has a named person next to it. Not "the marketing ops team" - a specific person who can answer questions about that feature. When the scoring model drifts, that person is accountable. When a consent management feature processes data in a way that needs investigating, that person handles it. When the quarterly review comes around, that person reports on whether the feature is still performing as intended. Ownership doesn't mean that one person does everything. It means one person is responsible for knowing what's happening with that specific feature and escalating when something isn't right. The difference between "someone should look at this" and "Sarah owns this and she's looking at it" is the difference between governance that works and governance that doesn't. They review outputs, not just inputs. The most common governance approach is to check the data going into AI features - is the data clean, is consent current, are the fields populated correctly. That's necessary but insufficient. A well-governed team also reviews what comes out. They pull a monthly sample of AI-scored leads and check whether the scores correlate with actual conversion. They review AI-driven suppression decisions to verify contacts are being excluded for valid reasons. They spot-check AI-generated content recommendations to ensure they're relevant and on-brand. They compare AI-assisted campaign performance against a baseline to see whether the AI is actually improving outcomes or just adding complexity. Output review is where you catch drift - the slow degradation in AI performance that happens as data changes, business conditions shift, and models age without recalibration. Input governance keeps the AI fed properly. Output governance keeps the AI honest. They tie review to the calendar, not to problems. Most teams only look at AI features when something goes wrong - when sales complains about lead quality, when deliverability drops unexpectedly, when someone notices an automation doing something it shouldn't. By then, the damage has been accumulating for weeks or months. A well-governed team reviews on a fixed cadence. The AI register gets reviewed quarterly. Scoring model performance gets checked monthly. Consent data gets reconciled annually at minimum and after any regulatory change. Platform upgrades get reviewed within a week of release to check for new AI features that may have been activated automatically. The cadence doesn't have to be aggressive. It has to be consistent. The difference between "we review when we remember" and "we review on the first Monday of every quarter" is enormous in practice. They document decisions, not just features. The register captures what AI features exist. Documentation captures why they were activated, what they're expected to achieve, and what criteria would trigger a review or deactivation. This matters because people leave. The person who activated a feature six months ago may not be on the team when questions arise. If the only record is "predictive scoring is on," nobody knows why it was turned on, what it was supposed to improve, or how to evaluate whether it's working. If the record says "predictive scoring activated in March to improve MQL-to-opportunity conversion, baseline conversion rate was 18%, target is 25%, review after 90 days," anyone on the team can evaluate the feature's performance and decide whether it should continue. This documentation takes five minutes per feature and saves hours of investigation later. It's the governance equivalent of code comments - nobody wants to write them, everyone is grateful when they exist. The real-world difference The practical difference between a governed and ungoverned AI environment shows up in specific moments. When a platform upgrade ships new AI features, the ungoverned team discovers them months later by accident. The governed team reviews the release notes within a week and documents any changes. When lead quality declines, the ungoverned team spends weeks investigating campaign creative, messaging, and targeting before someone thinks to check the scoring model. The governed team checks scoring model output as a first step because it's on the monthly review calendar. When a regulator or enterprise customer asks "what automated decisions does your marketing platform make?" the ungoverned team spends days trying to reconstruct an answer. The governed team opens the register and provides it immediately. When a team member leaves, the ungoverned team loses institutional knowledge about what AI features are running and why. The governed team has documentation that survives personnel changes. None of these scenarios are hypothetical. They're the moments where governance either earns its keep or reveals its absence. It's less work than you think The most common objection to operational AI governance is that it's too much work for an already-stretched team. In reality, the ongoing maintenance is minimal: Updating the register when features change - minutes per update. Monthly output review for scoring and key automations - one to two hours. Quarterly register review - one hour. Annual consent reconciliation - half a day. Platform upgrade review - one hour per release. Total: roughly 2-3 hours per month plus a half-day annually. That's the cost of knowing what your AI is doing. The cost of not knowing is measured in compliance incidents, degraded performance, and the hours spent investigating problems that proper monitoring would have caught weeks earlier. The teams doing this well aren't spending more time on governance. They're spending less time on firefighting. That's the trade-off - and it's one that every marketing ops team should be making.

Your company has an AI policy. It went through legal, got presented at an all-hands, and it's sitting on the intranet right now saying all the right things about responsible use, data protection, and human oversight. Meanwhile, someone on the marketing team activated a predictive scoring feature during last quarter's platform upgrade because it looked useful. It's been running ever since - deciding which leads get prioritized and which get buried - and nobody logged it, nobody checked what data it's feeding on, and nobody was asked to own it. The policy on the intranet has no idea it exists. That gap - between the policy and what's actually happening - is the difference between having AI governance and having AI discipline. One is a document. The other is a daily practice. Almost every company has the document. Almost none have the practice. The policy-to-practice gap There's a reason AI policies don't translate into AI governance, and it's not laziness or incompetence. It's structural. AI policies are written by legal, compliance, or senior leadership teams. They're written at the principle level - "use AI responsibly," "ensure transparency," "maintain human oversight." These principles are correct and necessary. They're also too abstract to guide the person in marketing ops who just got asked to activate an AI feature in the MAP. What does "ensure transparency" mean when you're configuring a predictive scoring model? Document it? Tell someone? Write it down somewhere? Where? In what format? Who needs to know? The policy doesn't say, because the policy was written for the board deck, not for the platform administrator. The result is a two-tier system. At the top, the policy describes how AI should be governed. At the operational level, AI gets deployed however the team sees fit - because nobody translated the principles into procedures that apply to the actual work. The policy says "human oversight." In practice, nobody is overseeing anything because nobody was told that oversight was their job. Five things organizations get wrong These patterns show up repeatedly. They're not edge cases - they're the norm. Writing the policy and calling it done. The most common failure. The policy gets published and the organization treats governance as complete. Nobody builds the operational layer underneath - the inventory, the ownership assignments, the monitoring, the review cadence. The policy becomes a compliance artifact rather than an operational tool. When something goes wrong, the organization can point to the policy and say "we had governance." But the policy didn't prevent anything because it was never connected to the systems it was supposed to govern. Governing the AI they chose but not the AI that arrived. Most governance frameworks cover deliberate AI deployments - the chatbot the team decided to build, the model the data science team trained. They don't cover the AI that showed up without anyone asking for it: the predictive features that came with a platform upgrade, the AI-assisted tools that a vendor enabled during onboarding, the smart capabilities that an individual team member activated because they saw them in a release note. In most enterprise marketing automation environments, more AI features are running by accident than by design. Governance that only covers intentional deployments misses most of the AI that's actually operating. Assigning governance to a committee instead of an owner. Committees review, discuss, and advise. They don't operate. When AI governance is owned by a committee that meets monthly, the AI features running inside the MAP get reviewed 12 times a year at most - and only if someone remembers to add them to the agenda. Operational governance needs named individuals with specific accountability: this person owns this AI feature, this person reviews its output, this person gets notified when something changes. A committee can oversee the programme. It can't run it. Monitoring inputs but not outputs. Many organizations focus their governance on what goes into AI - data quality, training data provenance, consent records. That's important. But it's only half the picture. The outputs matter just as much: what decisions is the AI making? Are the leads it scores converting? Are the contacts it suppresses the right ones? Is the content it recommends relevant? Output monitoring is where you catch drift, degradation, and failure. Without it, you're trusting the AI because you trust the inputs - and as we've covered elsewhere, those inputs may not be as reliable as you think. Treating governance as a one-time exercise. AI doesn't stay static. Models drift as data changes. Platform features get updated. Business conditions evolve. A governance framework built for the AI environment that existed six months ago may not cover the AI environment that exists today. Governance needs a review cadence - quarterly at minimum - that checks whether the inventory is current, the owners are still in role, the monitoring is still catching what it should, and the AI features are still producing the outcomes they were activated for. What discipline looks like in practice AI discipline isn't a framework or a programme. It's a set of habits built into how the team operates. When someone activates an AI feature, it gets logged - what it does, what data it uses, who turned it on, who owns it. That takes five minutes and it's the difference between a governed environment and a mystery. When the platform ships an update that includes new AI capabilities, someone reviews what changed and documents whether any new features were activated. That's a quarterly task, timed to release cycles. When AI-scored leads are handed to sales, someone checks monthly whether the scores correlate with actual conversion. If they don't, the model gets reviewed. That's not a project - it's a standing agenda item. When consent data changes - new regulations, updated processing purposes, revised preference categories - someone checks whether the AI features that consume consent data are still operating within the updated boundaries. That's an annual task at minimum. None of this requires new technology. None of it requires a dedicated governance team. It requires the same operational discipline that the best marketing ops teams already apply to their platforms - extended to cover the AI features running inside them. The discipline gap is the real governance gap The industry has reached a point where most organizations have policies and most organizations have AI running in production. The gap between those two facts is discipline - the operational habits that connect what the policy says to what the AI actually does. Closing that gap isn't expensive, complicated, or time-consuming. It's the kind of work that gets skipped because it's not urgent, not visible, and not rewarded - until something goes wrong and everyone wishes it had been done. The organizations that build AI discipline now will operate with a level of confidence and control that their competitors - the ones relying on policy documents and hope - simply won't have. And when the regulatory questions arrive, and the compliance audits happen, and the customer incidents occur, the disciplined organizations won't need to scramble. They'll already have the answers. The policy is the starting line. Discipline is the race. Most organisations are still standing at the start.