The question every marketer dreads
From the early days of my career, I still remember my first budget meeting where the CFO (who seemed so important, busy and downright scary for a green-horn marketer) leaned back in his chair and asked me, very politely, “So what did we actually get out of that?”
That was a silver sponsorship for a large and important trade show. Six figures, three days, a booth I was very proud of, and a stack of badge scans I was even prouder of (the team did so well!). A lot of work went into the event and even more at the event as we navigated through 5 high-touch activations on and off the tradeshow floor.
We were visible and the audience feedback was fantastic! We even created 2 weeks of video content to use on our social afterwards! What I didn’t have was an answer he could put in a spreadsheet immediately after the event – because not enough time had passed from when the event took place to when the question was asked. And events are a long-term game taking 12-18 months to show attribution.
If you’ve been in B2B marketing for more than five minutes, you’ve had some version of that meeting. And if you haven’t yet, just brace yourself – it’s coming.
That moment is why I’ve spent a good chunk of my career obsessed with attribution. It’s also why I think the conversation about attribution is about to change completely. So grab a coffee. Let’s talk about where we’ve been, where we are, and where I think this whole thing is going.
Why attribution has always been so darn hard
Here’s the dirty little secret nobody puts on the conference slide: attribution is, at its core, a data problem… long before it’s a marketing reporting conundrum.
Picking a model (or asking the MOps community: “Hey, what is the best model to use?”) is the easy part. We even get to enjoy arguing about first touch versus W-shaped versus custom weighted pretending like we are solving existential questions of marketing here. I’ve sat through those debates, and I’ve started those debates – all with the goal of bringing more awareness to the choices that marketing has to make before reporting this out with diligence and rigor.
What I never said out loud in those discussions is that the model sits on top of a very tall, very wobbly stack, and most of the work lives underneath it. So yes, let’s talk about all of the wobbly Jenga blocks in this stack because to get attribution right, you have to:
- Clean the data. Fix duplicate leads, junk form fills, enrich missing data points and half-filled records, deal with sales reps who log every call as “Other.” The old adage holds – garbage in, beautiful dashboards of garbage out.
- Structure it. Your CRM, your MAP, your ad platforms, your web analytics and your intent providers all structure and pass data differently, speaking slightly different dialects. Time series data (my favorite) vs. summary statistics data rolled up at 7 days, 30 days and 90 days; person-level or account-level (or people rolled into accounts), all fields or only some fields… All of this comes into play when somebody has to build a model where these systems’ output can actually be ingested and used for attribution.
- Stitch it. Tie anonymous website visits to known people, people to accounts, accounts to opportunities, and opportunities to revenue. Across months. Across buying committees of seven, ten, sometimes fifteen humans.
- Pick and tune the model. Then decide whether it reflects how your buyers actually buy, or just how you wish they did. Do you want to keep email touches and give them weight in your model? Do you choose to count impressions or drop them? How about social reactions? Ad clicks? You see, it gets complicated. Choices, choices…
- Separate signal from noise. That webinar registration from a competitor’s intern? Noise. The three pricing page visits from a VP at a target account? Very much signal. Knowing which is which takes judgment, and judgment takes context.
- Interpret it. A number without a story (and relevant context) doesn’t get you budget. A story without a number (and preferably an open opportunity that resulted out of the story) doesn’t either.
It has been 2.5 decades of why marketers couldn’t get attribution right – even in the heydays of digital transformation. And the core reason for that was that most marketing teams weren’t built with data engineers, SQL wizards and data scientists on staff. Demand gen and revenue marketers like me were hired to build brands and fill pipeline, and somewhere along the way we were suddenly expected to wrangle a warehouse. I took SQL Coursera lessons and Tableau classes to keep up. Many didn’t. But even with this background, I never tried to pretend I could do what a data analyst / data scientist or a data engineer could.
That gap is exactly why I fell in love with CaliberMind (and came to work for them) – it solved my existential problem of data wrangling that I never had the time or skills to fully address. Cleaning, structuring and stitching the data had to be handled for you, so marketers could spend their time on the part they’re actually good at: making smart decisions.
Attribution is now the price of marketing admission to the GTM table
A few years ago, having solid attribution made you look like a rock star. You walked into a planning meeting with a clear view of which channels drove pipeline, and the finance team looked at you like you’d brought donuts. Well, not quite – but as close to it as possible.
That era is over. Attribution is now table stakes.
I talk to a lot of growth marketers, functional revenue leaders, CMOs and marketing ops leaders, and I hear the same refrain over and over: the CFO (or Sr. Finance Manager, their right hand with a lot of veto authority) won’t even start the budget conversation until marketing can show what each channel returned – the full shebang: cost per opportunity, influenced pipeline, sourced revenue, time to close by segment, etc. If you can’t show your work, you don’t get a seat at the table (or a yes on your budget ask). You get something like “Let’s regroup, pull some more data and meet again a week from now”… But a week from now is only 2 weeks from the end of the quarter, and you need a decision now…
I would put a poker face on and say to myself : “Fair! Every other function in the business has to justify its spend, so Marketing doesn’t get a pass because our job involves creativity and vibes.” The same is true for many others.
By the sound of market chatter right now, here’s where a lot of teams are right now: they’ve finally gotten their attribution house in order, they can defend last year’s budget, and they’re feeling pretty good about it. But it is no longer the end of the story. Because now, the next question is: “OK, great. But what then?” Next steps, next decision, next action from the insight – what are those things? It is no longer speed to lead; it is the speed of insight to the execution of the next steps.
The marketing goalposts move again
The question leadership is asking marketers today sounds nothing like the one I got in that trade show meeting.
It used to be: “What did we get out of this?” And if we got something, you, the marketer, get permission to go and do the same thing again with more budget.
Now it’s: “How are you using AI to grow revenue five times faster?” After all, every business is AI-forward with a massive AI subscription bill. But very few are getting anything meaningful in terms of revenue augmentation out of it.
And so it goes, the marketing goalpost moves again. Now towards AI enablement and agentic activations.
It is true that looking backward and explaining what worked is still required. It’s just no longer enough, and it is the marketing expectations floor, a point of parity – no longer a point of difference. Boards and CEOs want marketing to look forward – like, wayyyy forward. They want to know how you’ll take everything you’ve learned from your attribution data and turn it into more pipeline, faster, with the same (or a smaller) team.
This is where I see a lot of us marketers get stuck. We built a beautiful attribution report on what audience engagements netted in terms of pipeline. We have an AI tool or three. We have a sales team with its own set of signals (or, they come to us for signals – at least in orgs where CaliberMind is the center of GTM intelligence). What these teams could benefit from – and what they don’t have today – is anything connecting those pieces. The report gets reviewed in a monthly meeting, someone says “fascinating,” and then everyone goes back to running the same campaigns they ran last quarter.
Insights that sit in a dashboard don’t necessarily get to move revenue. But an insight-driven action does! So, let’s talk more about insight to action – and beyond.
My version of what good marketing looks like
Let me paint you a picture, the one I think about on long flights.
Let’s say your attribution model is running on clean, stitched data. It knows which channels, campaigns and touches are moving deals forward, by segment and by persona. So far, so good. Full stop – and a victory lap. Marketing won! Uhm, not yet.
Now imagine this: AI picks up that output and reads it in real time. It spots that mid-market fintech accounts who engage with a certain type of content convert twice as fast when a CFO persona gets a specific message within a week. It marries that with the rest of your GTM intelligence: intent data, firmographics, product usage, where the account sits in the buying journey, what sales is hearing on calls.
Then it doesn’t just tell you (but please, do tell me – the human in the loop is so important in case I need an audit!). It does something about it.
An agent goes and builds the audience. It picks the right content, the right message, the right channel and the right moment for each persona on the buying committee. It drops that into a workflow your team already uses, whether that’s an ad audience, a nurture track, or a nudge to an AE saying “this account is heating up, here’s why, here’s what to say.”
And here’s my favorite part. Every one of those activations creates new engagement data. That data flows right back into attribution, the model gets smarter, the next round of activations gets sharper, and the whole thing keeps going. A continuous loop of learn, act, measure, repeat.
This is where the cycle of monthly reports that go nowhere stops. No more guessing which accounts to prioritize. No more “let’s revisit this next quarter.”
That is definitely the future of B2B marketing made possible by the AI infrastructure for the future at the foundation of marketing engagement.
Where we’re taking CaliberMind
Naturally, I’m not writing this from a mountaintop – there is a rhyme and a reason of where this is coming from. I’m writing it from inside the product roadmap, and I’ll say it as it is: our shipping schedule has never been busier.
Here’s what that looks like in practice.
Activations. CaliberMind already does the heavy lifting on clean, stitched data and multi-touch attribution. Now we’re closing the gap between insight and action. You can sync CaliberMind lists to LinkedIn, Google Ads and 150+ other destinations, as a one-time push or as a dynamic segment that refreshes on a schedule. With our LinkedIn Ads activations, your LinkedIn audience can mirror your CaliberMind segment exactly, so accounts and contacts drop in and out automatically as they qualify. No more exporting CSVs at 11pm the night before a campaign launch. (I’ve done it. We’ve all done it.)
Agent Cal. You can ask Cal to build those lists for you in plain English. That’s the first taste of what AI-driven activation feels like when it sits on top of data you actually trust.
LinkedIn as a signal engine. Our own data shows LinkedIn ad budgets growing at five times the rate of Google spend, and paid search conversion rates heading the wrong way. B2B marketers are voting with their wallets. Our technology partnership with LinkedIn brings even more of that engagement data into CaliberMind, which means more signals for AI to learn from, and a much richer picture of how buying committees move.
More muscle behind the vision. CaliberMind is now part of Integrate. We keep operating as a standalone product, with more resources, more execution channels and deeper AI capabilities behind us. Our shared goal is simple to say and hard to build (well, we are not afraid of “hard”): bring intelligent activations to every B2B marketer.
Put those pieces together and you can start to see the loop take shape. Attribution tells you what’s working. AI reads it, combines it with the rest of your GTM intelligence, and turns it into activations. Those activations generate fresh signals. The signals make attribution smarter. Round and round we go.
So, the next time a CFO leans back and asks what we got out of that, I don’t want to just have the answer. I want to tell them what we’re already doing about it.
