What is revenue attribution and why is it so important?

Once you’ve hit the $1M ARR mark and built out your go-to-market teams your organisation will most likely be deriving revenue from a mix of sources. These vary a lot, but might include a Sales Development Rep (SDR) team, salespeople sourcing deals themselves, partnerships, events, paid ads or content marketing. Being overly-focused on one source usually means that you're leaving opportunity on the table, and so it makes sense to see a bunch of sources at companies that have achieved some scale.

Diversifying your demand generation is great for broadening the top of your funnel, but it does add additional complexity to your go-to-market engine and make things more difficult to manage.

One of the tricky problems this introduces is revenue attribution. Once you have more than one source of revenue, you'll need to start deciding how to give sources "credit" for revenue. There are a few different methodologies you can use here, and each will have a different impact on the mental model you need to use when thinking about your customer acquisition. Let’s step through some options and how to think about selecting the right option for you.

Multiple, partial, and single-source attribution

Customer journeys are messy. While in theory we might think of the process as linear, the reality is that it is much more complicated.

Customers interact with us on their timeline, not ours. They also interact with us in a variety of ways, not just through one channel. As we can see in the winding path above, this hypothetical customer went through our organic, paid and SDR channels pre-purchase.

So if a customer attends one of your events in June, subsequently clicks on a paid ad in July, and then makes a purchase worth $800 in August who gets credit? You have three main options for attribution here.

Depending on the option you choose, you'll need to adapt your mental model for how you think about your customer acquisition engine. This is definitely a situation where the right approach is context-dependent, but regardless of where you end up you should take a stand on what the right approach is for your organisation. 

Making a decision

There are several factors in making the right decision for your organisation. 

The first is how complex your typical customer journey is and how much overlap there is between your demand generation channels. After all, if customers for the most part only interact with you through one channel pre-purchase, you’ll be defaulting to Single Credit attribution most of the time anyway.

The second is to what extent responsibility for different channels sits across teams. One of the key reasons attribution is important is in being able to tie the initiatives of a person or team to revenue output. If your marketing team owns demand generated through the website but your product growth team owns the generation of demand within the product, then multiple attribution becomes more compelling.

The third is to what extent attributed revenue ties to how an individual or team’s performance is measured. To give an example, Sales Development teams are often compensated on the basis of their contributed revenue. In this sort of situation, high fidelity around the total contributed revenue becomes very important and having multiple source attribution will likely be necessary.

The final factor is your ability to implement a solution in terms of data quality and technical capability. Oftentimes, the data simply doesn’t exist for you to be able to do certain types of attribution correctly. Even if it does, pulling it off can require complex data modelling work that you may lack the capability to perform.

Another way to think about the right choice for your startup is to walk through some examples - here are a few questions you’ll need to think through differently depending on the attribution methodology you’ve selected.

How much revenue did I generate through each of my channels this month?

Your choice of attribution strategy's most obvious impact is on how much credit various sources get for booked revenue. For example, let's say you acquired 5 new customers worth $100 each in the previous month with three from organic, one from paid and three from your Sales Development Rep (SDR) team.

Here's what attribution would look like based on this:

As you can see in the Multiple Source column, because some deals had both SDR activity and either paid or organic activity if we go with multiple source attribution we end up "double counting" some revenue across sources. This is the largest drawback of multiple source attribution.

The other two options will always sum up to the actual amount of revenue you closed, but will attribute revenue to different sources based on the particular methodology you're using. This can be equally tricky. To give an example - how are you going to compensate your SDR team? It's unlikely you're going to want to remove deal credit for them simply because someone decided to click on an ad before the rep reached out and booked a meeting.

What is my most efficient lead source?

Your expectation for how much revenue each source should generate will flex based on your attribution methodology. In a multiple touch attribution system, your expectation for how much revenue any individual source needs to contribute will be higher than in the others.

This has knock-on implications for how you think about the efficiency of each of your sources. As you can see from the table below, your close rates and deal sizes will look very different for each of your sources, based on how you’re attributing deals to them. 

Your SDR channel looks 3 times more efficient (15% close rate vs. 5% close rate) when using multiple source attribution! Your attribution model has implications not just on how you think about chopping up revenue between your sources, but also how you measure efficiency throughout your funnel. 

This article is more specific than topics I’ve covered before here, but I hope it’s a useful and practical way to think about different methodologies for attributing revenue to different sources and the pros-and-cons of each. No methodology is perfect but so long as you bear in mind the implications of your chosen model and tailor it to fit your company’s situation you’ll be on the right track to understand where your customers are coming from.