Why Attributer isn’t multi-touch

Attributer is a first-touch attribution system, which means it captures and passes through information on the lead’s first visit to your website.

We occasionally get asked if it can do multi-touch attribution, such as capturing and passing through information on both the first visit and last visit to the website.

I wanted to write this article to outline why we believe first-touch attribution is the best approach for the vast majority of businesses out there (and subsequently why we decided to make Attributer a first-touch system).

There are essentially 3 reasons why first-touch attribution is a better approach for most businesses than multi-touch:

1. Multi-touch is complicated

Imagine you get a new lead come into your CRM and it says their first visit to your website was from a Google Ad campaign, and the last visit was from a Facebook Ad campaign.

What do you do with that information? How do you make decisions with it? Do you attribute this lead to your Google Ads or your Facebook Ads?

The general approach to this problem is to split the credit for the lead into two. But how do you split it? Do you do 50/50 so you give equal credit to each touch? Or do you give more credit to the first visit because it’s fundamentally how they found your business?

What about if they have other touches between the first visit and the last visit? Imagine they clicked one of your Instagram Ads somewhere in there. Shouldn’t that get some credit as well?

This is where you then start to get into various ‘multi-touch attribution models’ such as the ones below:

• U Shaped Model – This is where the first touch gets 40% of credit, the last touch gets 40% of credit and the touches in between split the other 20%
• W Shaped Model – This is where the first touch gets 30%, the middle touch gets 30% and the last touch gets 30%. Any other touches split the remaining 10%

Before you know it, things have gotten out of hand and you’re in a world of statistical modelling, ensuring you have statistical significance and other highly technical things. 

This is great for large businesses with full-time data scientists on staff, but the vast majority of SMB’s in the world just want to be able to say ‘I got 5 customers from Google Ads and 5 from my Facebook Ads’ and this is what simple first-touch attribution products like Attributer allows them to do.

2. Needs high volumes of customers to work

Because you are splitting the credit for customers & revenue across multiple sources, you actually need to be getting a high volume of new customers each week in order to have enough data.

If you’re like most small-medium businesses and you’re getting 5-10 new customers per week, you’re going to end up with a situation where your reports look something like this:

• Google Ads – 2.7 customers this week
• Facebook Ads – 2.5 customers this week
• Organic Search – 2.9 customers this week

Ultimately, how is this going to help you make better decisions to grow your business? What would you actually do with that information? Would you increase spend in Google Ads? Probably not because it only generated an additional 0.2 customers than Facebook Ads did.

Because you’re dividing the credit between multiple-touch points and you’re not getting thousands of new customers each week, you’ll likely just have these kinds of low, relatively similar numbers that won’t actually tell you anything actionable (I.e. It won’t clearly show that you’re getting more customers through Google than Facebook and therefore you should invest more in Google).

3. It’s expensive and time-consuming

Before founding Attributer, I worked at several technology companies that were worth multiple billions of dollars and I implemented multi-touch attribution several times, so I have first-hand experience with how expensive and time-consuming it is.

As an example, at one company the software we purchased to do multi-touch attribution cost $30,000 per year and it took multiple highly-qualified people several months to implement it.

And ultimately, this was time and money we spent on an initiative that didn’t directly increase customers & revenue. Obviously these tools can help you figure out what you need to do to increase customers & revenue, but for most businesses the time and money you spend on implementing them would be far better spent on actual marketing initiatives that directly move the needle (like more ad spend, producing more content, etc) than on implementing a backend analytics system.

In conclusion

Based on the above factors, we felt that first-touch attribution was a better fit for the vast majority of businesses out there, and as a result, designed Attributer to capture and pass through first-touch attribution information (I.e. information about the lead’s first visit to your website)