Direct Load is not a Marketing Channel: Attributing to the Right Marketing Drivers
Companies that still use Last Click attribution tend to have a big bucket of revenue attributed to Direct-to-Site (DTS) or Direct Load. The people classified as DTS usually have great conversion rates. Fantastic! So how do we find more of those people?
First, let’s take a look at what DTS really means. Marketers treat DTS as a marketing channel, but it should be viewed as navigation. Typing a website or brand name into a web browser is the digital equivalent Direct Load is not a Marketing Channel: Attributing to the Right Marketing Driversof driving to a store. The customer intended to go to your site.
Branded Navigation Includes Branded SEM and SEO
Changes in browsers and search engines over the years make Direct Load, Branded Paid Search, and Branded SEO almost identical in the consumer’s eyes. They look the same on the screen and have similar conversion rates. In a sample size of one, I’ve watched my husband navigate to different sites using all three avenues. He probably didn’t think there was a difference with any of the three (though his purchase could have shown up in three different channels in a marketer’s report!)
So what? How do I use this to increase sales?
Back to the original question – how do I increase DTS buyers? If you’re still asking that question, you’re asking the wrong question. What you really want to know is what drove that person to go to the site. If you don’t know the drivers, you can’t take action.
We’ve thought about this problem a lot. You need to distribute DTS and branded navigation to understand the full effects of marketing. We take a two-step approach to this. The first step is a “raw” allocation that shows what drives a purchase at the individual customer level. The second step is a “full” allocation that distributes the Direct-to-Site and Branded Search to the marketing activities that preceded the branded navigation to the site. The two-step process is important because the allocations are not rules-based; the models need to learn the customer interactions and purchase drivers in order to understand what drives Direct Load so it can redistribute it accurately.
The simplified example below shows the two-step process for a $100 purchase. The first step looks at all the variables to understand what caused someone to purchase. In the first step, the Direct Load is included. In the second step, the full allocation, the $38.56 that was originally credited to Direct Load is distributed to the other marketing that preceded the purchase, giving Direct Mail a substantial lift. In this particular example, Customer Driven receives credit because of the prior purchase activity.
Without this second step, the marketing effectiveness for Direct Mail is dramatically understated. A marketer would make different mailing decisions knowing that her direct mail piece drove online sales that were originally credited to Direct Load. This also illustrates the need to measure all of a customer’s interactions, both online and offline, to understand how offline marketing impacts online orders, as well as how online marketing drives store sales.