I’m going to be brutally honest with you. Attribution of leads to sales is broken.
There’s a myth that gets peddled by naive marketers and believed by naive business people, that with the advent of digital, you can now measure everything. This is patently false. I hear it come through on social media, at events, from other marketers who should know better, and it frustrates me enormously. We have to start by acknowledging that attribution is substantially more complex than we want it to be. Digital has made it more complex, not less.
Why attribution is genuinely this hard
No one makes a serious purchase, and by serious I mean something more significant than a chocolate bar at a petrol station, based on one factor. People make purchasing decisions through many factors. The more expensive and complex the product, the more touchpoints they’ll need before they act. If you’re buying something complex and expensive, you need time and information, and research says that generally means somewhere between 7 and 14 touchpoints, mixed across online and offline.
In a digital world, we can measure very little offline activity. And even online, there’s a multitude of touchpoints that don’t talk to each other. Let me walk you through a realistic journey.
You hear about a new medical device from a friend who’s a healthcare professional. You forget about it for a while. Then you see something posted on LinkedIn about that product, which sparks a bit more interest. Some more time goes by, and you need to see the product several more times before you’re ready to act. Eventually you’re in a meeting where someone raises the need for exactly this kind of solution. You mention you’ve had some thoughts on it, and you jump onto ChatGPT to research your options. It gives you a range, one of which is the brand you’re already vaguely aware of. You go to the website to validate what ChatGPT told you, read the clinical case study, verify the information, and you’re ready to make a recommendation. A few days later, you Google the brand name and book a sales call.
In this scenario, the only data anyone captures is the ChatGPT referral and the Google search. Nobody connects the conversation with the friend. Nobody connects the LinkedIn exposure. All the brand awareness work, the combination of word of mouth and social content that actually built this person’s trust over time, is invisible. Instead, Google gets the credit. Someone in the business says great, we need to do more SEO. AI optimisation gets ignored, LinkedIn gets ignored, and both get defunded, despite being the things that brought that sales call in the first place. It’s a classic attribution issue.
Where the tracking breaks down further
Once that visitor becomes a lead, the problem doesn’t stop. Frankly, they might just ring you, and that call doesn’t get properly recorded either.
When you ask how they heard about you, they might say Google. They might say ChatGPT. They might say LinkedIn. They might say a friend recommended you. All of these are true. None of them are true on their own.
Then there’s what happens once the lead is actually recorded. In a lot of businesses I’ve seen, CRMs aren’t set up correctly in the first place. Leads aren’t tracked properly, they’re not converted into deals properly, and those deals aren’t tracked through to sales properly. That’s the scale of the problem.
So what can we actually do about it?
We need to acknowledge there are three types of marketing we need to be doing, and measuring, in parallel.
Brand awareness. Often carried by social media and other above-the-line activity, trade shows and the like. What we’re measuring here is eyeballs, the amount of exposure our brand is getting.
Engagement. How many people are engaging with our content on social media, engaging with informational content on our website, and what our traffic sources look like, including from large language models now, not just search.
Leads. Well, we know what this is, and hopefully they are being recorded.
The old marketing funnel. Invented in 1898, and still just as relevant as it ever was.
People are not simple beings. They make complex decisions in complex ways, and we can only measure what we can actually measure. Our job is to be saying the right things, at the right time, in the right places, to capture people as they move through that complexity.
The MedTech layer on top
In MedTech specifically, you potentially have three buyers, the clinician, the procurement people, and sometimes the end customer. You’ve got regulatory gates to clear. You may have an internal champion inside the company you’re selling to, who needs to sell on your behalf before you ever see a purchase order. Revenue can be a long way down the track.
This means marketing and sales need to be talking to each other constantly, at the macro-market level for marketing and the individual deal level for sales, about what’s working and what isn’t. If you’re only measuring revenue, you’re flying blind for most of the sales cycle. The question needs to be reframed from did this generate revenue, to how is our pipeline moving.
Splitting the pipeline
It helps to cut the pipeline in half. There’s the marketing pipeline, awareness moving into engagement, ending at the point a lead is generated. Then there’s the sales pipeline, which starts where the lead begins and runs through the sales cycle itself, varying depending on whether you’re selling to a clinician, a procurement team or an end user.
Leading indicators, not lagging ones
This is where the real discipline sits. A leading indicator predicts a later activity. A lagging indicator just confirms what already happened.
Early in the cycle, leading indicators look like clicks through to your website from LinkedIn or social content, engagement at a trade show, citations in large language models. Further down the pipeline, they look like clinical evidence and white paper downloads, references from existing customers, and identifying your champions mid-cycle.
None of this works if your CRM isn’t set up tightly enough to actually deliver the data. That’s not a nice-to-have; it’s the foundation everything else in this article depends on.
Reporting to leadership who want a number today
Here’s the real-world challenge. Leadership wants a simple ROI figure, now. If you’re eighteen months into a sales cycle, you often don’t have that figure in the early stages, and pretending otherwise helps nobody.
What you can do is link the steps together and make the pipeline visible, graphically, so people can actually see it moving. Build reports that talk to the whole pipeline, not just the end of it, so progress is visible well before revenue is.
Put simply, the companies that get MedTech marketing ROI right aren’t the ones with the cleverest attribution model. They’re the ones who defined what progress looked like before they started, tracked the process rigorously, and refined the execution at every stage along the way.
