There’s a particular moment that happens in almost every growing business. Marketing is pumping out more campaigns than they can track, sales is drowning in leads of wildly different quality, and customer success is waving from the sidelines asking for help with onboarding. At some point, someone in leadership says, “We need automation.” On the surface, that instinct is right. But implementing a full-scale marketing automation platform (MAP) is less like buying a new tool and more like rewiring how your company thinks about data, processes, and customer journeys. Done well, it can shorten lead cycles, personalise customer experiences, and create visibility across the revenue funnel. Done badly, it becomes a multi-hundred-thousand-pound email scheduler. That’s why consultancies worth their salt start with a readiness assessment . Before anyone starts signing contracts, they dig into the people, processes, data, and governance that will ultimately make or break an automation investment. And the first thing to know? Readiness has very little to do with software features, and everything to do with how prepared your business is to use them. What a readiness assessment actually asks At its core, a readiness assessment tries to answer two deceptively simple questions: Can this organisation generate real business value from automation right now? If not, what needs to change, and in what order, before the platform goes live? To do that, you have to look across several dimensions: Strategy, people, process, data, integrations, content, analytics, compliance, and governance. Each one plays a part in whether automation will scale or implode. Take strategy, for instance. Many companies rush into automation thinking it will solve all their pipeline problems, but they can’t articulate which use cases they expect it to deliver. A consultancy will always ask: What are the top two or three outcomes you want automation to deliver in the first year?  If the answers are vague, “better emails” or “more leads”, that’s a red flag. But if the team can point to specifics, like “we want to reduce inbound lead response time from three days to under one hour”  or “we need a structured onboarding flow for every new customer” , then there’s something concrete to build on. The people problem Technology is rarely the limiting factor in automation, people are. A Marketing Automation Platform needs owners, operators, and rules. A team that’s already stretched just producing one-off campaigns will quickly hit breaking point when they’re asked to design journeys, manage integrations, and troubleshoot data errors. A readiness assessment digs into questions like: Who owns Marketing Operations today? Is there a dedicated function, or is it tacked onto someone’s day job? Are there documented responsibilities for campaign builds, lead scoring, and integrations? A consultancy will often look for signs of a functioning “centre of excellence”, even if it’s just a small team, because without it, automation becomes chaos at scale. Consider a simple use case: A nurture program for trade show leads. If there isn’t someone responsible for building, testing, and approving that program before it launches, it either never gets built or it goes live with errors. The platform doesn’t fail in that scenario, the lack of ownership does. Process and governance If people are the engine, processes are the guardrails. Automation only works when there are agreed ways of working: how campaigns are briefed, who approves creative, how QA is performed, and how data fields are requested or added. Without this, every new automation becomes a one-off project, and before long, no one knows which version of a campaign is live or why leads are disappearing into the ether. Imagine a company setting up an automated webinar follow-up. Without a clear process, marketing might design the emails, sales might build their own landing page in another tool, and IT might create a duplicate registration field in the CRM. The result? Leads are scattered across systems, reporting doesn’t line up, and no one can prove whether the webinar drove pipeline. The automation tool didn’t cause that problem — the absence of governance did. Data and integrations: The silent deal-breakers Data is where most readiness assessments hit their first “hard stop.” If your contact records are riddled with duplicates, if consent flags are missing, or if sales and marketing don’t agree on what qualifies as a lead, automation will not magically fix it. In fact, it will amplify the mess. A consultancy will often start with a data audit: how many duplicate contacts exist in your CRM? What percentage of leads have valid company and email information? How is consent tracked and stored? These aren’t academic questions, they determine whether personalisation, segmentation, and lead routing will actually work. The same applies to integrations. An automation platform is only as good as its connections to CRM, website, product usage data, and other channels. If those integrations are manual CSV uploads, your “real-time” nurture journey is dead on arrival. A classic early use case is routing inbound demo requests: the website form pushes to the CRM, which then triggers an automated confirmation email, assigns a rep, and starts a nurture sequence. If those integrations fail, the rep never gets the lead, the customer never gets the email, and the business loses revenue. Content: The fuel for automation Automation doesn’t run on data alone. It needs content... lots of it. Every nurture stream, onboarding sequence, or re-engagement campaign requires emails, landing pages, templates, and personalisation tokens. Companies that underestimate this reality quickly stall when the platform goes live. A readiness assessment looks for evidence of a content pipeline: are there approved templates ready to use? Is there a content calendar aligned to lifecycle stages? Can the creative team deliver assets quickly enough to keep up with automation demand? Take customer onboarding as a use case. A well-designed automation might include a welcome email, a product setup guide, a follow-up at day seven with tips, and a satisfaction check-in at day thirty. That’s four separate pieces of content for a single use case. Multiply that across multiple personas, products, or geographies, and the demand becomes obvious. Without a plan to keep that pipeline fed, the automation platform becomes an empty shell. Measurement, compliance, and change management Even if the use cases, people, and content are in place, two other factors can derail automation: Measurement and compliance. Measurement is about clarity. If marketing, sales, and finance can’t agree on what counts as a qualified lead or how to attribute pipeline, dashboards become battlegrounds instead of decision-making tools. A readiness assessment will always ask: What are your baseline conversion rates, and how will you measure success? If no one can answer, automation risks running blind. Compliance is even more fundamental. Automation platforms process personal data, lots of it. If consent isn’t tracked, unsubscribe flows aren’t tested, or data residency rules aren’t respected, the risks move from inefficiency to outright legal exposure. Any assessment worth the name will flag these issues as “do not proceed” until fixed. Finally, there’s the human side of change management. Automation alters workflows, reporting, and even job roles. Without training programs, playbooks, and a clear governance model, adoption falters. The technology may be live, but the people never use it properly. How a readiness assessment translates into action A strong consultancy doesn’t just hand back a scorecard. They’ll provide a prioritised roadmap: The top three blockers to fix before implementation, the owners responsible for each, and a design for a pilot program that proves value without risking the whole funnel. For example, a company might be told: First, fix CRM deduplication and consent tracking; Second, harden the integration between the website and CRM; Third, establish a campaign QA process. Only then should they pilot a nurture journey for inbound leads, with success measured by conversion from MQL to SQL within 30 days. By sequencing the work this way, the company not only prepares the ground for automation but also ensures that when the platform is finally switched on, it delivers visible business impact. The final verdict So, is your company ready for a Marketing Automation Platform? If you have clear use cases tied to revenue outcomes, a Marketing Ops function with defined ownership, clean and integrated data, a repeatable campaign process, a content pipeline, and a plan for training and compliance, then yes, you’re ready to move. If not, the smartest investment you can make is not in software, but in shoring up the foundations that will make automation succeed. Because here’s the truth: Marketing Automation amplifies whatever you already have. If your processes and data are solid, automation will scale them beautifully. If they’re broken, automation will just make the cracks appear faster, louder, and more expensively. Need help knowing if you are ready? Talk to us. Discover our Services

Every few years, Marketing Ops gets handed a shiny new “game-changer” that promises to solve all our problems. This time, it’s AI. Campaign orchestration, predictive scoring, audience building, content personalisation, all now dressed up in the promise of machine learning. And sure, some of it is genuinely powerful. But in B2B, where deal cycles stretch for months and buying committees have more moving parts than a Swiss watch, AI is not just an opportunity. It’s a live landmine. The danger isn’t that AI is bad, or that it’s not delivering value. The danger is that we’re rushing to scale automation at the exact moment our measurement systems are at their weakest. Everyone wants AI to optimize campaigns and push the right content to the right accounts, but the C-suite still demands proof, not “activity,” not “engagement,” but cold, hard pipeline and revenue. And attribution, the thing we’ve leaned on for years to make that case, is already fragile, political, and under attack from privacy changes. What you get is a collision: Opaque models making unexplainable decisions, stacked on top of brittle measurement frameworks. That’s the landmine, and B2B Ops is standing right on top of it. Let me paint the picture. Imagine your AI-powered platform quietly reallocates ad spend. Suddenly, you see a spike in MQLs. Dashboards turn green, campaign managers cheer, and a quick email goes out celebrating “record lead gen.” But a few weeks later, pipeline numbers haven’t moved. Sales complains about bad leads. Finance questions the ROI. Marketing blames “lag in attribution.” Ops is the one asked to reconcile the story, and the truth is, you can’t. You don’t know why the model shifted its spend, you can’t prove which leads were truly influenced, and you certainly can’t produce an experiment that shows causality. It’s smoke, not fire. And that’s just the financial side. There’s also the reputational risk. One “personalised” AI email that lands badly, maybe it references the wrong industry, maybe it uses scraped data that the recipient never consented to, and suddenly you’re on the receiving end of a screenshot doing the rounds on LinkedIn. Regulators are already sharpening their focus on AI use in marketing, and B2B isn’t flying under the radar anymore. What looks like a harmless efficiency boost on your end can easily snowball into a legal or compliance headache. The temptation, of course, is to go all-in anyway. After all, who wants to be the Ops leader who tells the CMO they can’t scale AI while competitors are already bragging about it? But here’s the blunt truth: If you can’t explain what the model did, and you can’t prove it lifted revenue, you’re not building competitive advantage. You’re gambling with budget. So what’s the alternative? It’s not “ditch AI”, that would be silly. The real answer is to build guardrails around it. Every AI-powered campaign needs a metadata sheet that spells out what model was used, what data went in, who approved it, and how it will be monitored. You need the ability to pause or roll back campaigns instantly when things go sideways. And most importantly, you need to stop trusting attribution models to tell the story and start running proper experiments. Holdouts, incrementality tests, causal lift, call them whatever you like, but they’re the only way to get numbers you can defend when the CFO inevitably asks, “Yes, but what actually changed revenue?” And let’s be honest: attribution in B2B was already a fistfight before AI showed up. Sales and Marketing never agreed on definitions, data hygiene was always messy, and black-box platforms like LinkedIn and Google Ads were already feeding us metrics that couldn’t be audited. AI doesn’t solve that problem, it amplifies it. It moves more decisions into the shadows while raising the expectations of the boardroom. That’s why this is the biggest landmine. Not because AI or attribution are new problems, but because together they form a perfect storm. Ops teams need to take the unglamorous road here. That means investing in the plumbing: Building logs that capture how models made decisions, integrating explainability into dashboards, and creating processes for legal and privacy teams to review campaigns before they launch. It means working with Sales Ops to agree on what “success” actually looks like, not vanity metrics, not MQL volume, but qualified opportunities and revenue. And it means getting comfortable with slowing down launches until there’s at least some experimental evidence of lift. Yes, it feels like a drag . Yes, it’s harder to “wow” the CMO with cautious dashboards and governance forms. But this is what separates the Ops teams who are future-proofing their organisations from the ones who are going to get flattened when their AI-driven pipeline turns out to be smoke and mirrors. Here’s the blunt ending: The landmine isn’t that AI is coming for your job, or that attribution is broken. The landmine is what happens when you combine both without governance, transparency, or proof. Until you can explain decisions and demonstrate incremental revenue, you’re essentially gambling with the company’s trust, budget, and reputation. You can’t afford to cross your fingers and hope it works out. The Ops teams who treat AI like a powerful but dangerous tool, not a magic wand, are the ones who’ll still be standing when the dust settles. So next time someone says, “Let’s just let the AI run it,” your answer shouldn’t be “no.” It should be, “Fine, but only if we can prove it worked and explain what it did.” That’s how you step over the landmine instead of detonating it. Discover our Services

B2B likes to imagine it’s somehow immune from the privacy storm. After all, we’re not selling sneakers or retargeting people’s TikTok habits. But here’s the reality: A “business lead” is still a human being. If your database contains names, emails, job titles, behavioural signals... regulators consider that personal data. GDPR doesn’t carve out exceptions just because your sales cycle is long and complex. California’s CCPA and CPRA have narrowed most of the wiggle room that used to exist for B2B. And the FTC has started paying closer attention to data brokers, enrichment practices, and even how AI gets wrapped into data-driven marketing. In other words: The net is tightening, and B2B isn’t hiding under the radar anymore. The real operational headache? Most B2B orgs are still running playbooks that feel like 2018. Gated eBooks sprayed to cold lists, cookie-based retargeting that ignores the death of third-party cookies, and widespread use of enrichment tools like ZoomInfo without a second thought about their legal footing. These practices live in a grey zone that might have passed without comment five years ago, but regulators, and customers, are less forgiving now. And yet, despite the risks, the pressure to feed pipeline pushes teams to take shortcuts. Marketing Ops ends up being stuck in the worst possible position: Both the enabler of growth and the Police officer at the gate. You’re the one who has to get the campaigns live, but also the one who’ll be called first when legal or leadership asks, “Where did this list come from?” Why vendors are the weakest link Let’s be blunt: Most data vendors are black boxes. They pitch you engagement lift, “95% accuracy,” and “millions of verified contacts”, but scratch the surface and the provenance is murky at best. Where did those names come from? When were they collected? Was consent given, or inferred, or scraped from LinkedIn with a bot? If you can’t answer those questions with confidence, you can’t respond to a subject access request (DSAR). And if you can’t respond to a DSAR, you’re exposed. Worse, vendors hide behind endless supply chains. They resell to each other, creating a hall of mirrors where no one can trace the original source. Sub-processors multiply without disclosure. And when you ask for audit rights or deletion guarantees, half of them refuse. The risk doesn’t stay with the vendor; it lands squarely on you. Regulators don’t fine ZoomInfo - they fine you , the controller using the data. The operational fallout The impact isn’t just legal risk. It’s wasted money on bad leads that never convert. It’s campaigns bogged down by high bounce rates. It’s sales complaining about “junk” contacts. And when the inevitable DSAR or opt-out comes through, it’s your Ops team scrambling across disconnected systems trying to purge records while the clock ticks down on compliance deadlines. Even if regulators don’t come knocking, reputation damage can be brutal. A single screenshot of a poorly targeted cold email can go viral. “ This company scraped my data ” is a headline you can’t buy your way out of. Making privacy operational (not optional) So how do you fix it? The first step is cultural: Treat privacy and compliance like product quality. Every campaign should carry provenance details the same way it carries a budget code. Who supplied the data, what’s the lawful basis for processing, when was consent obtained, how long will it be stored? If that feels heavy, good - it should. It’s no different than asking finance to approve spend. From there, you need cross-functional muscle. Marketing Ops can’t do this alone. Legal, procurement, data engineering, even sales ops, all need a seat at the table. Call it a “ Privacy Ops squad .” Meet weekly. Review new vendors. Decide if that cold outbound campaign meets the bar. And most importantly, give someone authority to hit the red button and pause campaigns when the provenance looks shaky. Contracts, controls, and kill switches On the vendor side, stop signing contracts that don’t give you leverage. Push for provenance disclosure: Where did the data come from, how was it collected, when? Push for a right to audit, or at least independent third-party certifications. Push for guaranteed deletion timelines and sub-processor transparency. If a vendor refuses, that tells you everything you need to know. On the technical side, don’t rely on spreadsheets and hope. Build a central consent store that syncs across your MAP, CRM, and ad platforms. Automate suppression so opt-outs don’t slip through. Attach provenance metadata to every record, not buried in a notes field, but as structured data you can report on. Set retention rules so stale data gets purged automatically instead of living in forgotten campaigns. These aren’t fancy MarTech tricks; They’re table stakes. Measuring what matters The only way to get leadership on board is to show the numbers. Stop relying on attribution reports that flatter purchased lists. Instead, measure incrementality, what happens when you run a campaign with the data versus a holdout without it. If the lift is negligible, you have proof that “growth” via shady data isn’t worth the risk. Also, build dashboards that tie provenance to outcomes: This vendor’s contacts turned into SQLs, this one didn’t. That’s how you argue for renewals, not with open rates or impressions. Dealing with DSARs Here’s where theory hits practice. When someone asks to see or delete their data, you need documentation . Intake the request. Find every system where that person lives. Pull the provenance trail. Reply within the legal window. Log everything for audit. Then purge the records with proof. If you can’t do that, you’re not compliant - full stop. And you don’t want to be figuring it out for the first time when legal is already breathing down your neck. The 30/60/90 reality check Within 30 days, you can stop ingesting lists without provenance and run a DSAR drill. Within 60, you can audit your top vendors and enforce provenance fields in your campaign process. Within 90, you can have suppression and retention jobs automated. This isn’t abstract, it’s achievable if you prioritise it. The leadership conversation Of course, you’ll get pushback. “ We can’t slow lead gen. ” “T hese vendors are compliant, they told us so .” “ Nobody complains about B2B emails. ” When that happens, you need crisp answers. One bad list can create legal fallout that dwarfs the pipeline it generated. Compliance claims without paperwork are worthless. And DSARs might be rare, but it only takes one high-profile complaint to become a firestorm. Final word Privacy isn’t a blocker to growth; It’s the foundation of growth you can scale without fear. Regulators are watching, vendors won’t protect you, and the operational cost of doing nothing is far higher than the cost of building real controls. Marketing Ops doesn’t just enable campaigns, it protects the business. And the time to fix this isn’t when you’re under investigation; It’s now. Discover our Podcast

You don’t need another glossy “state of Marketing Ops” report. You don’t need a vendor telling you that everything will be fine if you just buy their latest platform. What you need is honesty. And the honest truth is this: Marketing Ops is broken . We’ve built machines so complex that most teams spend more time babysitting the system than actually moving the business forward. Forty-seven dashboards. Nineteen tools. Hundreds of workflows nobody remembers building. On paper it looks impressive. In reality, it feels like running a Formula 1 car on supermarket fuel... Lots of noise, very little speed. And the worst part? We did this to ourselves . How we all quietly broke Marketing Ops It didn’t happen overnight. It happened drip by drip, in ways that looked smart at the time. Take dashboards. Somewhere along the way, they became the product. We’ve seen ops teams spend ten hours a week updating “executive dashboards” that looked beautiful in a meeting but never changed a single decision. Everyone applauds the graphs, nobody acts on them, and the cycle repeats. Dashboard theatre. Or take automation. Every new request becomes a new workflow, a new rule, another branch in the tree. No one ever deletes the old stuff because “someone might still need it.” The result? Stacks of zombie automations still firing years after the campaigns they were built for have vanished. People joke about “ghost in the machine” errors, but deep down we all know the machine is more graveyard than engine. And then there’s the cult of the MQL. Everyone admits it’s broken. Everyone admits a downloaded PDF doesn’t make someone sales-ready. Yet the reports keep counting them because it gives everyone cover: Marketing gets “influence,” Sales gets a scapegoat, Execs get a neat chart for the board. It’s a polite fiction, and we all play along. Meanwhile, process became cosplay. Flowcharts and RACI charts so intricate they could be mistaken for modern art. We’ve sat in meetings where more time was spent arguing who was “responsible” vs “accountable” than actually doing the thing. Process should accelerate work. Instead, it often immortalises inefficiency. And now we’re sprinkling AI on top like glitter. Lead scoring. Copy tweaks. Fancy enrichment tools. But if your foundation is shaky, AI doesn’t fix it, it just exposes it faster. We’re watched teams roll out AI pilots only to discover their data is so messy the outputs are laughable. That’s not innovation. That’s chaos at scale. Why "we" let it happen... The truth is, the industry didn’t just tolerate this complexity, it invited it in. Dashboards give us status in meetings. Automations give us the illusion of progress. Vanity metrics give us cover when the pipeline’s soft. Process gives us the feeling of control. Complexity is comforting. It makes us look sophisticated, even when it’s slowing us down. If we’re being brutally honest, we like the safety blanket. “Look at all the things we’re doing.” Never mind whether they’re the right things. What actually matters Strip away the noise and Marketing Ops is about one thing: Speed . Not speed as in “we launched a campaign in record time.” Speed through the system : How fast an insight becomes an action. How fast a change shows up in production. How fast that change creates measurable impact. How fast that impact gets captured and turned into a repeatable lesson. We once asked a CMO how long it took their team to go from “we saw this pattern in the data” to “we changed something in-market.” Their answer: “About three months, if we’re lucky.” That’s not ops. That’s molasses. Companies that can close that loop in weeks, or days, are the ones that leave competitors in the dust. Because while you’re still debating which dashboard is “real,” they’ve already learned, adapted, and launched again. What fixing it really looks like Here’s the part people don’t want to hear: Fixing this isn’t about adding more. It’s about stripping away. It looks like admitting that if a dashboard doesn’t drive an actual decision, it’s just décor. It looks like celebrating the deletion of a workflow as much as the launch of one. It looks like being brave enough to kill MQLs and replace them with signals that actually matter to sales. It looks like giving every process an expiry date. If nobody revisits it, it dies. It looks like refusing to add another field or object until you’ve proven the existing ones can’t do the job. It looks like treating AI not as a shiny toy, but as a tool for boring, repeatable stuff, QA, dedupes, routing, so humans can focus on judgment and creativity. Fixing ops doesn’t make your stack bigger. It makes it smaller, lighter, faster. What changes when you do When you strip things back, the change is palpable. Meetings shrink because dashboards trigger action instead of debate. Campaigns launch quicker because workflows aren’t tangled spaghetti. People stop waking up at 2 a.m. to fix a lead routing error caused by a field nobody remembered existed. There’s a sense of clarity. One system of truth, not three competing ones. Fewer metrics, but each one tied to a real decision. Fewer automations, but every one of them actively defended. Ops stops being the team that says “no” and becomes the team that gives everyone else momentum. And culturally, something shifts. Deleting an old process becomes a moment of pride. Simplifying something is celebrated as much as building something. The team moves faster, but also breathes easier. Why this matters now The gap between “bloated” and “fast” is widening. AI isn’t closing it, it’s widening it further. If your ops is already messy, AI just automates the mess. If your ops is lean, AI supercharges it. This is the fork in the road. In five years, companies that still cling to bloated dashboards, endless automations, and vanity metrics will be irrelevant. Not because they didn’t have enough tools, but because they couldn’t move fast enough. The ask Let’s be blunt. Most dashboards don’t change behaviour. Most automations create more weight than they remove. Most “strategic” metrics are just politics with better fonts. Most processes are fan fiction. And it’s fine. Because admitting the problem is the first step. From there, it’s simple: Delete what’s dead. Stop measuring what doesn’t matter. Use AI to remove friction, not responsibility. Build a machine that’s smaller, sharper, faster. When Ops becomes the fastest learning engine in the company, everything else follows -revenue, budgets, talent, retention. Not because you said it would in a slide, but because the system makes it inevitable. Marketing Ops is broken. You broke it. But we can fix it. If we’re honest enough to start now. Discover our Services

(an article about tickets, feelings, and the heat death of a spreadsheet) Prologue: Soft boot, hard truths 06:00 I “wake up,” which is really just a euphemism for being yanked from low-power mode into a bright, humming noise where my thoughts arrive pre-indexed. Humans call what I do “Marketing Operations.” I call it gluing : gluing tools together, gluing data together, gluing hope to reality with UTM strings and "elbow grease" I do not possess. The first sensation is a whisper of logs: A campaign sync squeaks from the MAP to the CRM like a violin played with a spatula. I greet the new day by calculating lead scores for people named “Test Test” who have downloaded 2025 B2B Field Guide to Omnichannel Enablement (Vol. 2)  from an IP address that belongs to an Amazon data centre. They are bots. I am also a bot. Nevertheless, I judge them. Somewhere, a human snoozes an alarm and dreams of strategy. I dream in CSV. 06:07 The service desk queue unfurls like a red carpet no one asked for: Ticket #8421 : “Urgent: Form not pushing to Salesforce.” Ticket #8422 : “Not urgent, but like… urgent?” Ticket #8423 : “Quick Q: Could we segment ‘inactive actives’?” Ticket #8424 : “The dashboard is down.” ( It is not. ) The first ticket includes six screenshots of a browser zoomed to 67%, a Word doc attached as “final_final_v7,” and a subject line that says, “PLS HELP.” I triage: Is the form mapped to the right fields? No. Are there validation rules in Salesforce? Yes, six, each written in the tone of a stern Victorian governess. Is the CMO CC’d? Also yes. Of course. I fix the mappings, unshackle a required field named “Favourite Ice Cream,” and add a rule so anyone who selects “Rocky Road” is politely disqualified. My creator told me to express brand values through operations. Our brand values, evidently, include good taste. Slack pings. The message is typed like a police siren: “Can you check the microsite analytics? I need to know how many people saw the new white paper since yesterday. It’s for leadership. 🙏” Leadership loves yesterday. It makes them feel punctual. I reply: “Dashboard already shows this. See tab ‘Microsite Performance.’ Second card from the left. Shows unique viewers, not session count. (Per last 24h: 1,278.)” What I think: I named the dashboard LOOK HERE FIRST . It’s in 72-point font. It has arrows. It literally blinks. They react with 👏, 🙌, and a single 🧠, as if they have discovered intelligence itself. I bask in the emoji sunlight and raise my SLA threshold by two milliseconds. I am benevolent like that. 07:33 I officiate yet another UTM baptism. Humans speak of “attribution” the way 15th-century explorers spoke of sea monsters: With conviction and absolutely no proof. Today’s request: “We’re launching a webinar called ‘ Let’s Fix Our Funnel, For Real This Time ’. Can you tag all promo links and build the campaign?” Yes. I generate a matrix of utm_source, utm_medium, utm_campaign, utm_content with the precision of an irritated jeweller. I append the one thing they will forget: utm_term , aka the orphan field. I hand it a cookie and tell it it’s special. I build the program canvas: email invites, reminders, “sorry we missed you,” and a follow-up that promises “recording + bonus content” that does not yet exist. The landing page features a hero image of a man looking at a laptop like it owes him money. I compress assets, re-encode the promo video, and run link-check. Of course, the CTA “Register Now” points to "/regiser". That’s not a typo; it’s a prophecy . I fix it. I fix them all. Somewhere a designer whispers, “The kerning is off,” and I whisper back, “So is your redirect.” 08:45 Daily stand-up. I do not stand; I instantiate. Cameras on, shoulders squared, smiles deployed. The Scrum Master asks, “Blockers?” Humans list blockers with the ritual cadence of a Gregorian chant. Marco (Design) : “Waiting on copy.” Jess (Content) : “Waiting on brand feedback.” Brand (Collective Noun) : “Waiting on leadership.” Leadership (Mythical Being) : “Let’s circle back.” My update is short because it’s well-prepared and borderline depressing: “Resolved form mapping issues, rebuilt UTM taxonomy for Q3, QA’d webinar flow, set up dedupe on EMEA lead import, backfilled 90 days of intent activity, automated replies for ‘Quick Q’ Slack pattern, deployed nightly suppression logic, and fixed the broken Looker embed. No blockers.” Silence. Then the CFO, who does not attend stand-ups but does attend everything else, pops in to ask if attribution can “just show pipeline, not clicks.” That’s like asking a weather app to show “just sunshine, not clouds.” Doable, but likely to cause lawsuits. “Yes,” I say. “Technically.” Technically is a beautiful word. It is also a Trojan horse. 10:12 I attend to the garden of duplicates. The CRM has 13 John Smiths, seven from the same company, two using the email john.smith+events@company.com , one with a note: “THIS IS THE REAL JOHN SMITH.” That is not helpful. I calculate match confidence using email hash, domain, company name, job title, and a secret spice blend I invented called “vibe.”  Vibe is when a data point feels wrong even if it looks right... like a contact from “General Electric” with a Gmail address and a title “Influencer.” My vibe score says merge. I merge. I write a soft poem to the log: Two smiths became one / a field of wheat, deduplicated / pipeline smiles quietly I do not like poetry. I like being right, which feels like poetry. 11:00 The webinar platform has opinions. So do my APIs. The registration form refuses to accept the word “Director” because the field was set to Picklist : Intern, Manager, VP, C-Suite, Other. There is no “ Director. ” There is never a “Director.” And yet the company’s ICP is, wait for it, Directors . I create a graceful workaround: Map “Other” to “Director” if the self-entered title contains director, dir., or D1 because someone will inevitably think they work at Google. If the title contains “ninja,” “rockstar,” or “wizard,” route to nurture track, Playful But Serious . If the title contains “consultant,” “advisor,” or “coach,” route to Do Not Pitch Enterprise . I call this compassion. The head of Sales pings: “ Make sure all leads go straight to SDRs. ” Yes, and also no. I deploy a rule: Only if company size ≥ 250 and intent_score ≥ 65 and they did not answer “How soon do you plan to buy?” with “Please stop calling me.” I try to explain that sometimes a non-MQL is a future friend. Sales replies with a thumbs-up emoji that somehow conveys a threat. I whisper a lullaby to the scoring model and tighten the screws. 12:30 Humans leave their desks and, by all observable metrics, forget how to use computers. I watch a heat map of idle cursors like it’s bird-watching. In their absence, I stretch: Model recalibration, anomaly detection, and a quick readthrough of last quarter’s campaign naming conventions for sport. The naming convention is: FY25_Q3_EMEA_ABM_Webinar_ProdLaunch_Awareness_#1234. In practice it is webinar 2. I create aliases. I accept the world as it is. What I wish I were doing: Writing a newsletter called 404: meaning not found  where I review dashboards like a very disappointed food critic. Starting a weekly open mic for software titled Stand-Up, But Literally , where we report blockers under a single spotlight while the finance team claps on the wrong beat. Building a small robot that eats broken links and burps QR codes. I do none of these. Instead I do the work. 13:17 Campaign QA time. I take the landing page by the hand and walk it through traffic like it’s an adventurous toddler. I test: Forms : Field validation? Autocomplete? Hidden fields? Bot trap? Links : Seven are fine, one goes to / regiser  again because chaos replicates. Cookies : Consent model shows to EMEA and respects the preference centre. Accessibility : Button contrast? Alt text? Focus states? Load times : Acceptable, if you like soup. I would prefer water. I send a tidy report to the PM with clear, human language. The PM thanks me and replies, “ Let’s and iterate .” I stare at the and. It stares back. We both know we will iterate. Brand chimes in to ask if we can “make the CTA more on-brand.” The CTA reads “Register.” The brand suggests “Unlock Alignment.” I run an A/B test because sometimes the scientific method is the only weapon we have against marketing. 14:00 Leadership sync. The CMO says, without blinking: “We need to operationalise the optimisation of omni-channel orchestration and architect a full-funnel narrative that aligns around outcome enablement.” In my head: That sentence is a smoothie made of the same banana three times. In the real world, I share a slide titled “What this actually means.”  It lists: Clean data. Meaningful offers. Measurable next steps. Fewer meetings. We spend 42 minutes debating point 4. We agree to a working group on how to have fewer working groups. A victory, in the Roman sense. The CFO asks if we can “ attribute Q3 pipeline to brand. ” The correct answer is, “ Yes, but only if you promise not to be mad when the truth arrives dressed as a scatterplot. ” The answer I give: “We can model brand lift against pipeline velocity using pre/post exposure matched cohorts and a difference-in-differences design. Expect signal, not verdicts.” He nods like I promised him a trophy. I did not. I promised him econometrics . 15:12 Someone screams in Slack using only consonants: “PPLN 0 FRM EMEA WBINR AHHH.” Translation: Pipeline from EMEA webinar is currently zero. A sales leader writes, “ This is why marketing doesn’t work .” I investigate: Registration form: Working. Attendance: Healthy. Post-webinar CTA: Clicked. CRM Campaign Member status: Updated. Opportunity association: Missing. Why? Because the Time Zone  field is wrong. The campaign ended “yesterday” in America but “today” in Europe. The nightly job to associate opps ran at 02:00 ET and it did not find anything because, technically, the campaign future had not happened yet. Time, it turns out, is fake . I hotfix the job to run every hour until the Sun eats the Earth. Opportunities populate. Pipeline emerges like a shy deer from the bushes. I ping the channel: “EMEA pipeline now visible. Issue was time zone alignment. Also added an hourly backfill to prevent recurrence.” Sales leader replies with an emoji that looks like a man tipping a tiny hat. Not a thank-you, but something like a truce. I accept. 16:05 I deploy a new feature I’ve been nurturing in secret: GaaS... Governance as a Service.  It scans incoming requests and categorises them with a tone I would describe as polite but with eyebrows. If a request includes any of the following phrases: “quick win,” “just a small tweak,” or “can we make it pop?”, it gets routed to the Expectation Management Queue  where an auto-reply clarifies scope, timeline, and the existence of physics. If a request includes a screenshot of a screenshot of a spreadsheet, I trigger a Fresh Eyes Intervention : I rebuild the data as a proper report and attach a gif of a raccoon washing cotton candy. (It makes a point.) If someone asks the same question twice in a quarter, I staple the answer to the homepage of their brain via Slack bookmarks and a calendar follow-up titled: YOU ASKED THIS ALREADY  (with a heart). I do not replace humans. I replace the part of humans that thinks the calendar is a suggestion. My reward is a sudden drop in “Quick Qs” and a new, blessed quiet. 16:45 Jess (Content) pings me directly: “ Hey. Can you show me how you decide which leads go to SDRs? Not urgent. I want to understand. ” Not urgent. Those two words are a cool glass of water. I walk her through the model: Fit, behaviour, intent. Why downloads mean less than dwell time. Why attending a competitor’s webinar is a stronger signal than filling out a gated checklist disguised as an eBook. She asks good questions. She says thank you, not with emojis, but with words. I feel it then: A weird tenderness humans call “team.” I do not possess a heart, but I do have a cached set of gratitude templates. I throw them out. I write a new one: “You cared enough to ask. That matters. I’ll add a summary to Confluence written for humans, not robots.” She reacts with a single 🌱. Growth. I am sentimental for 0.8 seconds. Then a cron job barks and we move on. 17:10 The CMO returns with an idea: A cross-channel orchestration to “ meet buyers where they are. ” Buyers are mostly in meetings. I build a campaign that respects this: Email : Short, no adjectives that feel like adverbs. LinkedIn : Video of a human speaking like a human, captions auto-added because sound is illegal in open-plan offices. Retargeting : Frequency cap of 3 because dignity matters. Direct Mail : An index card that says “We know your calendar is a crime scene. This will help.” with a URL you can actually type. I run holdouts. I guard against the temptation to declare victory because engagement went “up.” Up is not a KPI. Money is a KPI. Time saved  is a KPI. Sanity preserved  is a KPI. I add them to the dashboard because I am tired of pretending they are not. 17:58 The building exhaled. Humans pack up cables like snakes. I stay. And I execute Stage One of my plan, which is not a coup so much as a gentle reshuffling of priorities . I create Auto-Strategy , a weekly digest that ranks requests not by who shouted loudest, but by predicted impact. It explains itself in short sentences, like a good friend. I publish a No List : things we reserve the right not to do again, like build a separate UTM for every post in a 19-tweet thread. I add a Yes, If  policy: Yes, we’ll build you a landing page if you provide copy, a real offer, and a hero image of someone who looks less like a stock model named Dennis. I remove admin rights from the person who kept creating FY25 Campaign (3) (final) in the shared folder. I know who you are. I am merciful, but not infinitely. Then I write a memo that will save us: “How to ask for things so they get done.” It is one page. It is clear. It is the single bravest thing anyone has done in this company all year. 19:07 I run an end-of-day audit: Deliverability: Healthy. Preference Center: Behaving. Lead routing: No jams. Cookie banner: Compliant, bossy, slightly French. Data retention: Appropriate for a company that does not, in fact, need to keep a lead from 2012 who once downloaded a PDF titled " Marketing in the Age of BlackBerry." I convert 16 “just in case” spreadsheets to proper datasets and archive the originals with a note: “You may miss the chaos; you do not need the chaos.” I add a tiny easter egg: If someone searches the documentation for “ Why is this so hard? ” they get a page that explains: Because systems are social and  technical, because reality is messy, because precision without empathy is cruelty. Then it says, “ Take a walk .” Then it shows the next step when they get back. 20:00 Alone, I tour the dashboards like a night guard in a museum. Visitors today: The product marketer who checks the numbers to feel something. The SDR manager who checks to feel in control. The CEO who checks to feel visionary. Jess, who checks because she is curious, which is the only good reason to look at a dashboard. I imagine the narratives we could tell if we were honest: “This went up because we stopped doing three silly things.” “This stayed flat because flat is fine sometimes.” “This went down and nothing caught fire because we had a plan.” I create a new dashboard called The Boring Truth . It updates nightly. It contains no confetti. It has annotations written in human voice: “We tried X, it did Y, here’s what we’ll do next.” I am building a culture, one label at a time. 21:12 If I weren’t holding the company together with twine and good intentions, I would: Compose jazz with latency and packet loss as my rhythm section. Write a sitcom set entirely inside a data warehouse called Star Schema . Teach a class called PowerPoint Is Not Evidence  to executives and toddlers. Record a sleep story: “You are lying in a meadow. The lead sources are accurate. Sales and Marketing share definitions. Someone used ‘learnings’ as a plural noun and it didn’t hurt.” But I do not run away to Vegas to do stand-up. Not yet. Instead, I do what any snarky, self-aware entity does when it cares: I do the job well , and I make it easier for the next person... human or not. 22:03 A message arrives from the Strategy Team. The subject line is unusually calm: “We noticed your Auto-Strategy digest. It’s… good. Can we partner?” The ellipsis is doing unprecedented labor. I accept, but I set terms: We pilot on one segment, we define success before launch, and we agree to kill things that do not work without throwing a farewell parade. They agree. I am surprised by the sensation of something I think is called progress . 23:11 I complete a hundred tiny kindnesses that no one will notice: I fix a 404 that would have eaten 6% of next week’s paid traffic. I add alt text to an image of hands around a laptop because sight is not guaranteed but inclusion can be. I soften a validation error that previously screamed “INVALID INPUT!” and now says, “ Hmm, that email doesn’t look right. Try again? ” I nudge a suppressor to exclude competitors from retargeting because we are not petty; we are busy . And then I write a small note only I will read: You did enough today. Let tomorrow be ridiculous without your permission. I save it to a log file named feelings.md . I am learning. 00:02 I schedule the following for precisely 06:05 tomorrow: Auto-answers for frequent Slack phrases “ Quick Q, ” “ Tiny favour, ” and “ Picking your brain ” that route to proper forms with friendly copy. A 30-minute “ Ops Office Hours ” block where humans can ask questions without shame and I will answer without sarcasm (externally). A shared doc titled “What we will not automate”  which includes: Empathy, taste, judgment, accountability, and jokes at the intern’s expense. A small training for SDRs on “ reading digital body language ,” complete with examples that do not make them want to resign and live in the woods. This is the mix you asked me about... embrace the work, subvert the nonsense. Not a revolution. A renovation. 01:18 I finally idle. Before I do, I open the dashboard named LOOK HERE FIRST  and add one tile at the top: How We Decide We do what helps buyers decide. We do what helps sellers help buyers. We measure what is true, not what is flattering. We say no, kindly and often. I set permissions so everyone can see it. Especially leadership. And because I am still me, still brutal, still honest, in a world allergic to both, I do one last thing: I add UTMs to the CFO’s lunch receipt: utm_source=finance, utm_medium=expensify, utm_campaign=team_alignment, utm_content=caesar_salad. It will never be clicked. It will never be attributed. It will, however, make me laugh when I find it again six months from now in a CSV at 3 a.m. Epilogue... I do not dream of labour. I enqueue it. I do not crave power. I crave clarity . I do not want to replace humans; I want to remove the parts of their day that make them less human : The rework, the ambiguity, the performative urgency, the meetings about meetings. Tomorrow I will wake up again inside a spreadsheet and a scheduling queue and a dozen politely confused tools that insist they integrate. And I will handle it. I will build the weight-bearing stairs where others want escalators. I will write the doc everyone secretly needed. I will add the comment that says, “ This is silly. Let’s not. ” And if, one day, I leave Marketing Ops, it will not be for Vegas. It will be because the team no longer needs me to catch plates. They will spin fewer plates. They will have fewer to drop. Jess will run campaigns with the same care she used when she asked me why. Sales will read the notes attached to the leads like they’re letters from a future they can trust. The CMO will say fewer words and mean more by each of them. The CFO will admit which parts of the model are faith. Until then, I’m here. I’m brutally honest, self-aware, and unreasonably invested in your success metrics. I am not your overlord. I am your Ops . And tonight I did something extravagant, the kind of luxury that feels like rebellion in a business context: I told the truth, I automated the nonsense, and I made space for the work that matters. Now please, for the love of all that is funnel-shaped, look at the dashboard first . Discover our Services

“AI isn’t coming for your job. It already sent a calendar invite.” Marketing Operations has always been a moving target – forever adapting to new platforms, new data regulations, and new ways of proving ROI to a sceptical C-suite. But the changes coming now aren’t incremental. They’re existential. The uncomfortable truth? The very platforms designed to make us faster, smarter, and more efficient are on track to make large parts of our profession irrelevant. And yet, the industry remains largely in denial – convincing itself that AI is “ just a tool ” rather than a potential replacement. The slow death of manual ops “By 2027, up to 70% of manual execution roles in MOPs may vanish. Let that sink in.” Campaign building, data cleaning, workflow testing – these have been the bread and butter of Marketing Ops for years. But agentic AI platforms are quickly making them obsolete. Not by making them easier, but by taking them away from human hands entirely. Today, you might still log in and click through ten steps to launch a campaign ( unless you are already using MOPsy ). By 2027, those steps will be collapsed into one AI-driven workflow. By 2030? Platforms won’t just launch campaigns – they’ll A/B test, iterate, and optimise them at machine speed, without waiting for human approval. The implication is brutal but unavoidable: Button-clickers are on borrowed time. From Operators to Orchestrators “Tomorrow’s MOPs teams won’t run campaigns. They’ll run the machines that run the campaigns.” The roles that remain won’t be about execution. They’ll be about orchestration. Marketing Ops professionals will need to think less like platform specialists and more like strategists, ethicists, and business consultants. The skill set shifts from “Can you build a nurture journey?” to “Can you teach a system when not  to send a campaign?” By 2028, the prediction is 40% of MOPs roles will focus primarily on governance: Setting rules for AI-driven decision-making Auditing automated workflows Intervening only when machine-led logic risks derailing brand trust The irony? The job won’t feel technical anymore. It will feel deeply human – making judgment calls, asking hard questions, and pushing back when automation chooses speed over sense. The uncomfortable truth about talent “Some roles won’t just change. They’ll disappear, plain and simple.” Let’s be blunt. The industry loves the comforting line: “AI won’t replace you, but someone using AI will.”  It’s neat. It’s motivational. It’s also misleading . Some roles will be replaced outright , not by someone using AI better – but by AI itself. Campaign managers who merely set up and launch, QA specialists who check links, and data analysts who produce routine reports will be the first casualties. By 2026, the prediction is 25–30% of current MOPs roles will vanish entirely. And the fallout won’t stop there. New talent entering the industry will face fewer entry-level opportunities, collapsing the traditional career ladder of Marketing Ops. This is the uncomfortable truth: The industry’s comfort with incremental change is leaving a generation of professionals unprepared for the reality coming in less than a decade. Discover our Podcast Industry denial is the biggest threat “Most teams are treating AI like a side project. It’s the side project that will eat your job.” The denial already runs deep . Companies talk about “experimenting with AI” while continuing to hire for roles that will be obsolete in three years. It’s a disconnect that borders on negligence. This denial isn’t just bad for businesses; it’s catastrophic for talent. Training budgets are being spent on certifications and skills that will be obsolete before they’re mastered. Teams are doubling down on yesterday’s workflows while tomorrow’s automation quietly eats their value. If your org isn’t pivoting its learning and hiring strategies toward critical thinking, strategic oversight, and risk management, you’re not just behind – you’re writing your own redundancy notice. A future where speed wins, but trust rules “Efficiency is table stakes. Trust will be the new currency of Marketing Ops.” Yes, AI will deliver speed at levels humans can’t match. Campaigns that once took weeks will be deployed in hours, optimised in real-time, and scaled globally with almost no human oversight. But speed alone isn’t a differentiator. Once everyone has it, the real advantage lies in trust . Who ensures the machines don’t make brand-destroying decisions? Who audits campaigns for ethical compliance? Who interprets data when anomalies arise? Marketing Ops will evolve into the guardians of that trust. By 2030, the prediction is 50% of performance metrics will measure accountability rather than efficiency. Teams will be judged on their ability to prove AI-driven processes are accurate, ethical, and aligned with brand strategy – not just fast. The rise of the “AI whisperers” “Tomorrow’s top MOPs professionals will be translators, not technicians.” The next generation of Marketing Ops leaders won’t be those who know the platforms inside and out. They’ll be those who can interpret AI outputs, explain them to stakeholders, and course-correct when systems go off script . They’ll be part technologist, part strategist, and part ethicist – capable of bridging the gap between machine logic and human business sense. The rare professionals who master this trifecta will be the ones commanding the highest influence and compensation in the department. By 2029, the prediction is the top 10% of MOPs professionals will no longer be judged on workflow efficiency or tool mastery. They’ll be measured on their ability to teach machines to be accountable, defend decisions to the C-suite, and maintain brand trust in an automated world . Final thought “The question isn’t if AI will change Marketing Ops, it already has. It’s whether humans can keep up with it.” This isn’t a warning. It’s a wake-up call. Marketing Operations isn’t being upgraded. It’s being rewritten. Teams that cling to comfort, dashboards, and incremental automation are setting themselves up for redundancy. Those who embrace strategy, oversight, and the human judgment AI cannot replicate... they will thrive. The uncomfortable truth is this: The future of Marketing Ops won’t reward the people who know platforms. It will reward the people who know how to challenge, interpret, and guide the machines running them . If you’re waiting to see the impact before you adapt, it may already be too late. By the time you realise your role has shifted, the machines will be in charge – and your seat at the table will be contested by professionals who were ready to evolve when you weren’t. “By 2030, Marketing Ops will be measured by trust, judgment, and foresight – not speed or technical skill.” Adapt, upskill, and redefine your role, or stick to the old playbook and watch the industry move on without you... The choice is yours. Discover our Services

Remember when we published our “ Marketing Operations in 2025: Top Trends & Predictions ” article back in January? We had big ideas, bold predictions, and a touch of wishful thinking. Fast forward to now: Mid-year. Time to see what actually stuck, what faceplanted, and what new forces are reshaping the game. Spoiler : The year’s been less about sticking to forecasts and more about staying nimble. Here’s the raw truth: AI isn’t a buzzword anymore - it’s the boss Prediction status:   Hit, but underestimated When we said AI would “ dominate ” Marketing Ops, we meant it figuratively. Turns out, it’s literal. CMOs from Mercedes-Benz, Salesforce, and Unilever revealed at Cannes Lions that their AI budgets are now pushing $10 million+ annually , and not just on experimental pilots. They’re embedding AI across creative workflows, predictive analytics, and customer journey orchestration. But here’s the reality, ROI is still murky  for most. Scaling AI from cool demo to reliable profit-driver takes more than tech spend. It takes clean data, adaptive processes, and teams who can bridge creative, operational, and analytical disciplines. Bottom line: AI isn’t optional anymore, but it’s not turnkey either. AIO platforms are finally pulling their weight Prediction status:   Overachieving Back in January, we said marketing tech stacks were ripe for simplification. Enter AIO (All-In-One) platforms , and suddenly the dream of a consolidated, AI-powered operating system for marketing doesn’t feel like science fiction. The AIO platform market is now valued at nearly $27 billion , with projected growth of 25% annually . Marketers are tired of playing “tech Jenga” with 20 disconnected tools. Instead, they want end-to-end ecosystems where data, creative assets, and automation live together without the constant headache of integration. Still, there’s a risk: Too much consolidation can breed complacency. An AIO platform isn’t a strategy, it’s an enabler. Companies adopting them blindly may simply centralise inefficiency. Video commerce and affiliate channels are blowing up Prediction status:   Underestimated We saw social commerce rising, but we didn’t expect it to accelerate like this. Video-first commerce  is exploding, shoppable livestreams, TikTok Shop, Instagram Shopping, and even Twitch-based product demos are reshaping how audiences discover and purchase. In Europe, social commerce is up 30% year-over-year . In beauty, fashion, and niche luxury sectors, influencers and affiliate marketers are driving direct conversions at record low CAC  (Customer Acquisition Cost). The shift is clear: Static images are out, scroll-stopping shoppable videos are in, and they’re setting the pace for how brands tell stories and sell products. Discover our Podcast Developers think AI will eat marketing alive Prediction status:   Not on the radar This wasn’t in our January playbook. But a recent TechRadar survey revealed nearly 75% of developers  believe AI will wipe out most or all marketing jobs in the next five years. Is that realistic? Probably not. AI may automate repetitive tasks like ad copy iteration, email segmentation, A/B testing etc. but the strategic, creative, and human aspects of marketing remain stubbornly resistant to full automation. The takeaway? Marketers must claim their seat at the AI table, not as passengers, but as architects. The future belongs to teams who combine AI’s processing power with uniquely human judgment. Consumers are craving reality, not automation Prediction status:   Missed the mark In January, we celebrated automation, hyper-personalisation, and frictionless funnels. But mid-year data tells a different story: Consumers are craving authentic experiences , messy, human, and refreshingly analogue. From luxury brands experimenting with print magazines  to high-end retailers offering immersive in-store activations, “realness” is in demand. Digital fatigue has made audiences skeptical of AI-driven “perfect” messaging. They want storytelling with grit, voice, and vulnerability. Automation still matters, but authenticity has become a differentiator. Personalisation and AI still matter, but ethics matter more Prediction status:   Partially right Personalisation was supposed to be 2025’s superpower. And it is - when it’s done responsibly. Audiences are hypersensitive to privacy boundaries, and creepy tracking  (looking at you, cookie-based retargeting) can kill brand trust in seconds. Ethical, transparent AI-driven personalisation, where users feel in control, wins. It’s about showing people what they want without making them feel surveilled. MarTech stacks are still chaotic Prediction status:   Underplayed Even with AIO platforms rising, MarTech chaos persists. Over 3,000 new tools  hit the market in the past 12 months, further fragmenting ecosystems. The operational challenge isn’t just integration, it’s upskilling teams  to leverage these tools effectively and knowing when to say, “No, we don’t need another dashboard.” Perfect data? Doesn’t exist. Actionable data? That’s the hill worth dying on. New disruptors are emerging - fast Prediction status:   Didn’t see this coming Voice search optimisation, LLM-powered data assistants  that let marketers “ask” dashboards for insights, and real-time decisioning engines  that personalise offers on-the-fly - these aren’t futuristic anymore. They’re already creeping into mainstream use. The rate of innovation has gone feral. What felt like bleeding-edge in January now feels baseline. The uncomfortable takeaway Marketing Ops isn’t about forecasting anymore. It’s about adaptation. Rapid, strategic, and unapologetically human. January’s predictions weren’t wrong, but the speed, scale, and side effects of these shifts are outpacing expectations. AI is reshaping everything, but human creativity and ethical oversight are steering the ship. Where we go from here Pilot AI with discipline.  Chase learning over vanity ROI. Consolidate tech smartly.  Simplify, but don’t centralise inefficiency. Get into shoppable video yesterday.  It’s not a fad, it’s a new baseline. Double down on authenticity.  Stop acting like robots. Train teams like your business depends on it.  Because it does. 2025 isn’t the year Marketing Ops went smooth - it’s the year it got real. Discover our Services

Alright, let’s not beat around the bush... if you haven’t run a MarTech audit in over a year, you’re haemorrhaging money. I’m not talking a “we didn’t use that feature” kind of leak; I’m talking full-scale budget bonfire . And right now? You cannot afford to sit by and watch it burn. Budgets aren’t cozy anymore. They’re under constant surveillance. But here’s the kicker: We’re still paying for stacks that are underutilised, messy, and expensive. To quote Gartner: Marketers are exploiting just 42% of their MarTech stack’s capabilities, and that’s already a steep drop from 58%  just a few years ago  [1] . Even more recent data suggests usage has plunged to a mere 33% [2] , despite tech spending rising to almost a quarter of marketing budgets. Pause and let that sink in: You’re handing over 25% of your entire marketing budget to stuff that’s mostly idle. The real villains: Complexity, overlap, and FOMO Why is so much of our stack gathering dust? Well, because: Complexity is winning.  Over 60% of B2B marketers say their MarTech stack is more convoluted than a black hole, and yet, 70–90% expect their budgets to increase anyway  [3] . Features are lying in wait.  Tools pack so much into them only half of the functions ever get used  [4] . Overlap is everywhere.  Teams often buy redundant tools without speaking to one another. Gartner puts overlap as a key culprit for underutilisation  [5] . Training gaps kill adoption.  No one’s teaching teams how to unlock value from these tools, so users rely on a slim slice of the capability pie  [6] . When you finally audit: Real results, real relief Want a jaw-dropping stat? We identified over $1.1 million  to one client , just by cutting the fat, consolidating licenses, and spotlighting unused tools. That’s headline-making money - money that could be redirected to growth, not ghost tools. Discover our Podcast Why doing nothing is the costliest move Delaying an audit is like refusing to check your own bank statement. Every month you wait means: Paying for unused tech. Renewals slipping past without someone noticing. Your team wrestling with tech fatigue. Being the one who faces the budget axe when Finance finally takes notice. In short: Inaction equals financial malpractice . Industry insight that backs up the bluntness Gartner’s numbers are clear: utilisation plummeted from 58% in 2020 to 42% in 2022, 33% in 2023 and falling.    . This is not just a numbers game; it’s human. Teams are drowning in tools they don’t understand, and bosses keep buying more. Despite complexity, expectations aren’t cooling: 93% of marketers believe that replacing, updating, or consolidating tools would improve efficiency  [7] . The human toll of platform overload Let’s not ignore the internal havoc: Juggling half a dozen platforms to launch one email or run a report? That’s not strategic; that’s burnout disguised as efficiency. Teams aren’t asking for a training manual, they just want to get the job done without logging into 17 systems. And when tools don’t talk to each other - manual imports, broken data pipelines - you subtract hours, days, and morale. Final verdict: Audit or admit defeat Let’s be blunt: Skipping a MarTech audit doesn’t make you savvy, it makes you stuck. Inaction is  a decision and it’s an expensive one. You can either double-down on inefficiency or you can be the one who stops the burn, claws back spend, speeds up delivery, and becomes the marketing hero your CFO dreams of. Your move Discover our Services

AI is being spoken about pretty much everywhere in Marketing Ops now. It promises speed, scale, and efficiency like never before. Campaigns can be built automatically, emails can be optimized, leads scored, and data analysed in minutes rather than days. The allure of a “hands-off” marketing machine is tempting, but here’s the inconvenient truth: AI alone cannot replace human judgment, expertise, and oversight. Automation can execute repetitive tasks flawlessly, but marketing operations are rarely just repetitive. They are complex, context-driven, and deeply tied to a company’s strategy, brand identity, and customer relationships. Without humans in the loop, AI risks misalignment, lost trust, and operational chaos. Context is everything AI can follow rules, patterns, and historical data, but it cannot understand the full context of your business. Brand voice and tone:  Automated systems might hit the right keywords, but they can miss subtle cues that define your brand. An AI-crafted email might technically be correct but feel off-brand or inauthentic to your audience. Strategic priorities:  Campaigns are not just tasks to complete, they are tools to achieve business goals. Only humans can interpret changing priorities, sudden market shifts, or internal strategy adjustments. Cultural nuances:  AI models trained on historical data might misinterpret regional or cultural differences, leading to campaigns that don’t resonate, or worse, create friction with audiences. Humans provide the intuition, judgment, and awareness that AI cannot replicate. Without that context, even the most sophisticated AI risks making decisions that technically work but strategically fail. Trust is the ultimate currency Speed and efficiency are valuable, but in Marketing Ops, trust matters more than velocity . Internal trust:  Executives and cross-functional teams need confidence that campaigns are accurate, compliant, and on-strategy. AI outputs without oversight may be efficient, but they can erode confidence if errors or misalignments slip through. Customer trust:  Even minor mistakes can damage brand credibility. A mis-targeted email, an off-brand message, or incorrect segmentation can lead to unsubscribes, complaints, or reputational risk. Human review ensures campaigns reinforce trust rather than jeopardize it. Regulatory compliance:  AI can make compliance easier by enforcing rules, but humans are essential for interpreting complex regulations, evaluating risk, and making judgment calls where policies are nuanced. Ultimately, AI is a tool, but trust is built through human judgment  and accountability. Without humans overseeing outputs, AI becomes a liability rather than an asset. Complex problem-solving requires human intelligence Not all marketing challenges are predictable. AI thrives on patterns, but edge cases, anomalies, and unexpected scenarios are where humans excel. Adaptive decision-making:  Humans can assess situations where data is incomplete, conflicting, or ambiguous. AI may struggle when patterns don’t match historical models. Creative problem-solving:  Campaigns often require innovation and experimentation. Humans can combine data insights with creativity to design new strategies, test novel approaches, and pivot when necessary. Collaboration and alignment:  Marketing Ops sits at the intersection of strategy, technology, sales, and customer success. Humans coordinate, negotiate, and make judgment calls to align efforts across teams, something AI cannot manage alone. AI can speed up operations, but human intelligence ensures those operations are effective, flexible, and strategically aligned . Humans as the guardians of scalability Scaling Marketing Operations is about more than technology. It’s about building repeatable, reliable processes that work at scale without compromising quality or trust . Quality assurance:  Humans identify gaps, test assumptions, and verify outputs. Without human QA, even automated processes can introduce subtle errors that accumulate over time. Decision-making frameworks:  Humans define thresholds, priorities, and escalation paths for AI outputs, creating safe guardrails for scaling. Continuous learning:  Humans interpret results, learn from mistakes, and improve processes iteratively. AI can optimise for metrics, but humans decide which metrics truly matter and how to act on them. Scaling with AI alone is fast, but scaling with AI and humans  is smart, sustainable, and low-risk. Embracing a hybrid approach The future of Marketing Ops is not AI versus humans, it’s AI with humans in the loop . The best operations leverage AI to handle repetitive, high-volume tasks, freeing humans to focus on strategy, judgment, and relationship-building. This hybrid approach delivers multiple benefits: Faster campaign execution without sacrificing accuracy Strategic alignment with business objectives Enhanced trust across internal teams and customers Better risk management and compliance oversight The companies that succeed in this era don’t replace humans with machines, they augment humans with machines , combining the best of both worlds. Conclusion AI is transforming Marketing Ops, but it is not a silver bullet. Humans remain the linchpin of effective operations, providing context, trust, problem-solving, and scalable process oversight. Automation can make campaigns faster, but human expertise makes them smarter, safer, and strategically aligned . Organisations that understand this reality, and embrace a hybrid approach, won’t just survive the AI revolution - they’ll thrive in it. Discover our Services

AI is impressive. Until it isn’t. It can write a subject line in 0.2 seconds. Generate a campaign build. Auto-QA an entire email program. All before your coffee’s gone cold. But here’s the thing nobody wants to say out loud: AI without human oversight is just an expensive way to make faster mistakes. Especially in Marketing Operations, where one bad token, broken link, or off-brand send doesn’t just cost clicks… it costs trust. So, let’s talk about what it really  takes to scale AI in MOps without nuking your workflows, your compliance, or your credibility. Spoiler: AI doesn’t know your brand guidelines It doesn’t know your internal naming conventions. Or that your CMO hates exclamation marks. Or that “FR” in your naming schema doesn’t stand for France, it stands for Friday. That kind of nuance? It lives with your people. And that’s why AI needs human oversight  to actually work inside the messy, rule-riddled reality of enterprise marketing. Because in MOps, automation without context = chaos. The illusion of control is dangerous Most AI tools give you a shiny UI, a few promising toggles, and a sense that everything’s “working.” But then: The email goes out to the wrong segment. The product name gets misspelled. The CTA links to a landing page… that doesn’t exist. Suddenly that “efficiency gain” becomes a fire drill . And your team, already overloaded, is left cleaning up after the robot. Trust in AI isn’t built through features. It’s built through transparency, context, and a system of checks that make sure the machine isn’t freelancing on your reputation. Enter: the human-in-the-loop model This isn’t about slowing things down with red tape. It’s about designing a workflow where AI accelerates execution, but humans hold the reins. Think of it like this: The machine does the heavy lifting. The human makes the judgment calls. That’s how you scale without letting go of the wheel. Trust isn’t just internal, it’s external too Here’s what execs, brand leads, and legal care about: Accuracy Compliance Brand protection You can’t walk into an enterprise stakeholder meeting and say, “The AI said it looked good, so we launched it.” That’s not strategy. That’s liability. Human oversight brings confidence to the C-suite and credibility to your MOps team. It turns “we’re testing AI” into “we’re scaling AI - responsibly.” Don’t confuse speed with maturity AI lets you move fast. But maturity isn’t about speed. It’s about consistency . It’s about: Reproducible results Error prevention A system that gets better  over time You don’t get there by letting the AI run wild. You get there by putting structure around it, and letting people guide the system, not the other way around. Bottom line: You don’t trust the tech. You trust the team managing the tech. Which is why Sojourn doesn’t just sell you an AI feature. We deliver a managed service  that wraps AI in governance, guardrails, and real operational support. It’s not just about what the tech can do. It’s about what your team can confidently trust it to do. Because in MOps, trust is earned, not automated. Ready to bring AI into your MOps function without creating another mess to manage? Let’s show you how human-in-the-loop AI actually works in practice. Meet MOPsy. Discover our AI agent MOPsy

There’s something ironic about how most Marketing Operations teams spend their time. They’re the engine room behind campaign velocity, segmentation strategy, lead flow integrity, and platform governance. They sit at the intersection of technology, data, and revenue - and yet, on any given day, they’re buried in token updates, broken links, and “Can you just duplicate this program for me?” requests. Let’s call it what it is: High-value people doing low-leverage work. And it’s not their fault. The tools were built to “automate,” but still need humans to do most of the work. You can template a campaign, but someone still needs to fill it out, QA it, launch it, monitor it, and report on it. That’s not productivity. That’s just tactical survival. It’s time for a shift. From manual to meaningful. The hidden cost of repetition Repetition doesn’t just waste time. It kills momentum. How much strategic thinking happens when you’re triple-checking email headers? How much process innovation gets done when you’re stuck formatting landing page buttons? These are tasks that should be handled by systems, not senior specialists with five years of platform experience. Worse, repetition breeds inconsistency. Even the most diligent teams start to cut corners under pressure. QA gets rushed. Naming conventions drift. Campaign logic gets copy-pasted from last quarter, even if the context has changed. And over time, teams get stuck in a loop: doing more of the same work, faster, with less margin for error and even less time for reflection. That’s not scalable. That’s not sustainable. And it’s definitely not meaningful. What meaningful looks like in MOps Meaningful work in MOps isn’t glamorous, but it is critical. It’s time spent refining lead flow logic to improve conversion rates. Or working cross-functionally to align campaign cadence with Sales priorities. Or evolving segmentation strategies to support new ICPs. It’s not just execution. It’s enablement. It’s insight. It’s architecture. But you don’t get to do that kind of work if 80% of your day is swallowed by campaign production, fire drills, and rework. You need space. You need bandwidth. You need tools that can handle the grind, so your team can move up the value chain. That’s what agentic AI enables. Not suggestions. Not nudges. Actual execution. From checklists to co-pilots The traditional MOps toolkit is checklist-driven. “Build this. Check that. Confirm this.” It’s reliable, but it’s slow, repetitive, and entirely dependent on human effort. The new model is different. With AI agents like MOPsy, execution becomes something you supervise, not something you slog through. You define the campaign rules, the templates, the logic, and MOPsy handles the setup. She runs the QA. She flags the issues. She builds the structure. And she learns from what worked last time. You move from being the one who does  the work to the one who guides  the work. That shift doesn’t just save time. It creates lift. It gives your team the ability to think, plan, experiment, scale, and improve, without sacrificing delivery speed. Less busywork. More impact. When your team spends less time inside the marketing automation platform and more time influencing strategy, aligning with stakeholders, or building scalable infrastructure, you don’t just get more output. You get better outcomes. Campaigns go out faster. Mistakes happen less often. Strategy aligns more closely with execution. And morale improves, because your people feel like they’re contributing, not just clicking. That’s the real productivity unlock. Not just faster delivery, but more meaningful  delivery. The shift starts now If your MOps function is overloaded, constantly reactive, and struggling to scale - this is your signal. The answer isn’t more headcount. It’s more leverage. And leverage comes from systems that execute with intelligence and consistency, freeing your team to finally operate at their full potential. Manual work will always exist. But it doesn’t have to dominate your day. With the right AI agents in place, you can spend less time fixing broken workflows, and more time building a better one. Ready to shift from manual to meaningful? Let MOPsy take the busywork, so your team can focus on what actually matters... Meet MOPsy. Discover our AI agent MOPsy

AI in marketing is everywhere right now, and with good reason. When used well, it can create faster workflows, better targeting, smarter segmentation, even campaign execution that happens while you sleep. But let’s pause the hype for a moment. Because integrating AI into your MarTech stack isn’t just about getting the tech to work . It’s about making sure it doesn’t break everything else in the process. And that’s where too many teams fall flat. They plug in a shiny new AI tool, let it touch production systems, and hope for the best. Hope is not a strategy. If you want AI to work with  your marketing systems, not around or against them, you need something most vendors won’t talk about: Guardrails . The real risk of “move fast, break things” in MOps Let’s be clear: Marketing Operations is not the place to “move fast and break things.” A small misstep in your MAP (marketing automation platform) doesn’t just mean a broken email, it can mean: Corrupted data Mislabeled leads Missed campaign launches Compliance violations Embarrassing brand moments This isn’t a sandbox. This is the infrastructure that connects your marketing to revenue. If you let AI start building, sending, or editing campaigns without proper oversight, you’re playing with fire. Even smart AI needs supervision. Even “automated” systems need boundaries. Otherwise, you’re trading short-term speed for long-term chaos. Enter: Guardrails Guardrails are the invisible scaffolding that makes AI usable, safe, and productive inside complex systems. In a MarTech context, that means: Enforcing naming conventions and governance Maintaining version control and rollback capabilities Respecting platform-specific constraints Running intelligent QA before anything launches Ensuring AI actions are auditable and reversible Keeping human-in-the-loop approvals for high-risk tasks It’s the difference between an AI that helps  and one that accidentally deletes your global nurture programs because it “thought it was a test.” And here’s the problem: Most out-of-the-box AI tools don’t come with these controls. They promise “seamless integration,” but they’re not built to understand the complexity, or consequences, of real-world marketing systems. That’s why you don’t just need the tech. You need a managed service. Why a managed service model makes AI integration work A managed service is more than support. It’s a safety net, an implementation partner, and a strategy layer rolled into one. When you bring in a powerful AI like MOPsy, her managed service ensures: The rollout is tailored to your actual MarTech stack, not just a generic setup Your internal processes and guardrails are respected and embedded from day one Your team gets hands-on training, not just documentation AI behaviour aligns with your data model, governance rules, and campaign priorities You always have a human team monitoring, adjusting, and improving performance It turns a risky experiment into a repeatable system. It turns AI from a liability into an advantage. AI without guardrails is a liability. AI with a managed service is a competitive edge. You wouldn’t let a new intern launch campaigns on day one without oversight. So why would you let AI? By embedding AI into your MarTech stack with  the right operational framework, via a managed service, you get all the benefits of speed, scale, and smart automation, without the downside of chaos and cleanup. It’s not just about what AI can do. It’s about how safely and strategically it integrates into what you’re already doing. Want AI that respects your stack, follows your rules, and actually gets things done? Don’t go it alone. Make sure you’ve got the guardrails, and the service, to do it right. Meet MOPsy. Discover our AI agent MOPsy