Attributing credit for customer conversion to the last ad clicked is unrealistic. Andrew Newman, director of operations and analytics for Europe at Rocket Fuel offers advice for correct attribution.
From the first pamphlets and billboards, to the latest promoted story on Facebook, marketers have always strived to prove a correlation between their advertisements and sales.
Although online advertising allows marketers to know when a customer has seen their ad, then clicked through to the point of sale, it doesn’t tell the whole story.
Online advertising does not exist in a vacuum, yet many advertisers and agencies still choose to assign full credit for a purchase to the last ad seen.
In a world of integrated marketing and mass media consumption, the notion that 100 per cent of a complex purchase decision, ingrained with sociological and psychological influences, can be attributed to a single advert, completely defies logic.
On the other hand, apportioning credit to every possible touchpoint your target may have come into contact with before he or she converts (a process often referred to as fractional attribution) is a complicated technique and is therefore seriously underused.
Of course, a far more realistic view is that conversion takes place after multiple exposures to multiple ad sources. While it may be simpler to give credit to the final click, logically we know this doesn’t tell the full story.
A broader approach to attributing success pays far greater dividends and delivers a truer picture of your channel ROI. In this economic climate, identifying where the best combination of value lies should be marketers’ top priority. While it may seem like a prohibitively complicated process, really it is about asking the right questions.
1. Do clicks measure success?
Before addressing the idea of last click attribution or fractional attribution, it’s important to make sure you’re not putting too much emphasis on clicks in general.
Some marketers remain convinced that ad clicks are the only actions guaranteed to lead to conversion. However, just as a person who has read a magazine advert might not act upon that influence straight away, online ad views that don’t lead to immediate clicks can, of course, still influence sales at a later point. The lead-time between exposure and action with print advertising and online ads are clearly different, but this example still illustrates that it is short-sighted to become too fixated on clicks.
2. Can we prove post-click ROI?
Time and time again, a strong correlation between a user converting and how recently he or she saw an ad is proven.
Based on this, marketers can draw curves to analyse the time lag between an ad impression and a purchase. If the ad works, it will work fast, generating more of the desired responses in a shorter timeframe – thus showing a steeper curve. From the angle of this curve, it is possible to predict post-click revenue.
3. Who gets the credit?
So, how should a display campaign distribute credit? Should credit be assigned to each of the digital initiatives that influenced that final action?
In short, yes. Value must be ascribed to partners throughout the funnel, even if the conversion didn’t take place during that first session. The reason for this is that the top of the funnel helps build brand awareness, while the mid-funnel guides the customer to the next stage and then on to conversion.
Collecting user data across channels through the use of a single technology enables marketers to review all touchpoints for each user on the path to conversion. Use pixels to help track who saw what advert, and then follow the customer journey after exposure.
This data is vital in identifying which partners are most effective at different parts of the funnel, allowing marketers to play to their strengths.
Some partners may be very strong influencers for example, though they are not credited with the last touch prior to the conversion.
4. What is the best mix?
Before we can use all this data to optimise our campaigns, we need to make sure we have all the information we need. Conversion data must be comprehensively tracked, we must also note how much revenue comes in and from where, and differentiate between returning customers and new customers.
It’s absolutely essential to track bottom-line results, usually sales, for more informed models. From here, campaigns can be updated and adjusted ‘in-flight’ based on changing data.
Ultimately, the best way to select our digital mix is through experimenting with different channels and partners. It’s likely the same bottom-line results can be more efficiently achieved with fewer partners and fewer touchpoints. The key is determining which partners and touchpoints they should be.
Marketing efforts should be integrated throughout the sales cycle, reinforcing each other as much as possible. All areas need to be ascribed value and optimised in conjunction with one another. This is called a ‘full-funnel’ approach.
Some marketers might argue that it’s too complex, but with the right data at your fingertips it doesn’t have to be so hard. The technology to run tests, place pixels in the appropriate places, and gain insight into who views and is influenced by your ad is at our disposal. Why not seize it?
