Beyond the Stats: Attribution and the Buying Journey
As a Client Strategy Manager working with national and emerging brands on their Amazon business, I’ve found that one of the key factors separating the highest growing sellers from the rest of the pack is their understanding of consumer behaviour and how that behaviour affects strategic decisions.
The biggest mistake I see businesses make is developing tunnel vision on their KPIs — especially regarding return and margin at the top of the funnel. When you’re a small business, pursuing efficient sales at the bottom of the funnel is a great place to start, but for brands looking to create a category or be the dominant player in a channel, you have to read between the lines.
To do that, you need to understand the attribution models being used in your channels, and how those models describe the actual behaviour your customers exhibit en route to a purchase.
Let’s look at the most common attribution model across marketing channels: last click. In this model, the last ad that a user clicks is given full credit for the order, as long as that click happened within the attribution window (typically 7, 14 or 30 days). This model is probably the most simple way to look at customer behaviour and enables more sellers to dip their toes into digital marketing. But it also forces marketers to put more focus on the lower funnel — where customers are closest to making a purchase.
Here’s an example to illustrate a typical consumer journey. Because it’s easiest to speak from experience, I’ll use the example of someone searching for a product to help with their dog’s fear of storms…
The “customer journey” of this person doesn’t begin until a particularly bad storm comes through, and their dog keeps them up all night. Exhausted the next morning, they search for “dog calming” and “fear of thunder in dogs” on Amazon to learn what products might be available to help.
In doing so, they’ll be presented with a pretty standard search results page on the Amazon mobile app, which consists of two ads and one organic listing above the fold. They’ll click into a couple and get a feel for the options available, evaluate if they even think they’ll work, and most likely close out to go about their day and consider it more in the future.
If it doesn’t storm for a while, they may even completely forget about their search. But the next time a storm comes though, they go straight to searching for the product type that stuck with them most: “dog calming treats.” Here, perhaps weeks later, they’ll find another ad at the top of their results for one of the products they clicked on previously and decide to make a purchase.
From a pure attribution perspective, all of the money spent on the click for “dog calming” was completely wasted while “dog calming treats” had a great return — leading the advertiser to stop delivering ads for the former while bidding heavily on the latter.
But what if the user had never seen or clicked on the calming treats in the first search? If the three listings above the fold didn’t even feature calming treats, would the second search have happened at all? It doesn’t seem likely, since research tells us that most users never make it past the first three listings on the SERP.
Too often, I see brands decide that the initial searches, at the top of the funnel, aren’t worth pursuing. They then wonder why their total sales begin to decline while they’re focused on the bottom.
When you consider that most buying journeys span days or weeks and involve as many as a dozen touch points across multiple channels before a purchase — which is still most likely to actually happen at a brick and mortar store — you might begin to see how murky the waters of your KPIs really are.
If you’re a retailer looking to poach a few sales from the bottom of the funnel as profitably as you can, Amazon’s 7-day, last-click model in Seller Central is probably a great fit for your business needs. But if you’re a brand looking to scale, which most are, then you’ll need to build out robust reporting that enables you to read between the lines.
As a starting point, here are some things to consider:
- Do you know what the average customer journey looks like in your category? If you sell, say, charging cables, the purchasing process could be minutes. But larger, big ticket items may take weeks or months. If you’re unsure about your consumers’ journeys, analytics platforms like Google Analytics can help shed some light.
- Does your reporting track spend, ad sales, and total sales in a way that you can view against one another? Further, are you able to track spend aimed at the top of the funnel against results at the bottom?
- Once you have these pieces together, look for the trailing change in other areas as you increase spend at the top of the funnel. If your customer’s journey lasts two weeks, do you see an increase in lower funnel or total sales after that period of increased spend at the top?
Reaching users early in the discovery phase creates secondary searches and builds your brand equity. In turn, that awareness drives down the cost of all of your advertising, regardless of the stage you’re reaching customers in.
In order to scale, you have to come to a holistic understanding of your customer’s purchasing path and what the attribution models you work with are actually telling you about it.