Attribution is overrated

I was talking with the founder of Syft Data about marketing attribution. Imran has invested years in building a better way to identify leads both to act on them and to attribute marketing success.

I realized halfway through the conversation that attribution wasn’t the real problem. It’s just not that important. Of course, it’s really hard. I agree with that. But focusing on attribution is misleading.

Effective GTM has to be anchored in the actual customer's journey. Today those journeys are complicated. Understanding what led to a purchase seems like a critical input to making decisions about marketing–but that doesn’t mean focusing on attribution.

Attribution might be hard, but just because it's hard doesn't make it as important as some would have you think.

The hardest part of marketing

The hardest aspect of marketing–and the most important–is getting the attention of your customers. Whether that's for something entirely new, standing out as a differentiated product in a crowded field, or not standing out but having people pick your product even if you are undifferentiated, this is essential for building a company.

These are all different challenges and require very different approaches. OpenAI or Cluely are great examples of marketing something truly new and novel. OpenAI created a whole new way to interact with computers and solve problems. Cluely got its start when a college student built an app to cheat on an Amazon interview with hidden AI–and got the offer (but declined it). It has grown through viral, controversial marketing. Cluely overlays AI on your screen so that you can “cheat” at meetings, which to them simply means perform better. It’s a new use case–and needs brilliant marketing to succeed.

Skyp is an example of a product executing on a differentiation strategy in a large, crowded market. There are so many emailing tools I still learn about new ones–with tens of thousands of customers, doing $10mm or even $50mm in ARR. Our marketing is meant to define why a company should choose Skyp. In a big space, there's always room for one more competitor if it can differentiate.

Booking.com is one of the few tech companies that entered a market as an undifferentiated competitor on purpose and dominated it. Their thesis: nobody cares where they book–what matters is appearing high in SEO and having an optimized checkout flow. This creates a powerful growth loop in a competitive market (structurally lower CAC, more profit, poured back into more acquisition). The consumer doesn't realize they booked on booking.com until much later. They just clicked a well-placed ad and flowed through a streamlined checkout. For Booking's direct-response model, attribution is absolutely necessary–they need to know which ads work. But most B2B sales motions aren't simple funnels. Trying to simplify them into "Clicked ad → purchased" risks disaster.

When channels work, nobody really cares about attribution. Think about big law firms doing hundreds of millions in revenue per year. Do you think the partnership ever discussed last-click attribution? No. They talked about whether the TSR3 is better than the G430 driver. Because the clients just came. They were differentiated. Things were working.

What you don't do is sometimes more important

It is very tempting to focus on attribution as a problem worth solving. But what if you don’t? I would rather have an educated guess at what is working so I can spend my time doing more of it than waste time trying to be very, very sure what I thought was working is in fact working.