By: Brian McAllister, VP, Media Operations, NetSeer
Let’s say you’re an advertiser running 1,000 online ads, and I told you that 999 of those ads were a complete waste of time because they weren’t converting. You’d reply that all 1,000 ads are valuable because they’re all hitting consumers during different points of the purchase funnel.
Then why are you still using last-click attribution, which pins all value on that .1% conversion rate? You’d have better returns playing the lottery!
Alas, last-click attribution is extremely inefficient. It also leads vendors to prioritize serving that last ad to users at the lowest possible rate right before they convert – at the expense of all the other ads along the way that have influenced the purchase decision.
The same goes for last-click attribution’s close cousin – last-view attribution.
And, since the most common online activity for most people is checking emails or using Facebook, guess what? Those are the vehicles where most people happen to make their last click or view before they convert!
Last-click or last-view attribution, of course, can also lead to manipulation and gaming of the system in order to get that final click or view. But good luck to those unscrupulous vendors! Because the truth is that the vendor who serves the last-click or last-view ad will always be, well, largely a matter of luck.
Vendors are either buying from open public exchanges, direct from sites, or from trading desks. But we’re essentially bidding on the same inventory at certain rates, just with different algorithms.
If Netseer happens to get credit for the last click or view, but the previous 10 clicks or views were served by Rocket Fuel, is that really fair?
Vendors don’t operate in silos completely independent from each other. The pieces of the online ad ecosystem are interconnected in an interactive web of their own. At every stage of the funnel process, we’re working together to drive users to one thing: conversions.
So bold brands and their agencies need to start changing their attribution models. Don’t give 100% to the last click or view – maybe give it 20%, give the first one 30%, and spread the remaining 50% over all those clicks or views in-between.
Every brand – indeed every campaign – needs to have a different formula for this. Some formulas will favor the upper funnel, some the lower funnel, for instance.
And advertisers can‘t just base their payments on where ads were served, but on the contextual relevance of those pages. If the last ad was viewed or clicked via an email, for example. maybe give it less credit
The rise of mobile and cross-device use further complicates the building of these new formulas.
And anything so complicated scares off marketers. Regardless of how inefficient last-click and last-view attribution are, it’s easier to just keep using the old system.
Somebody has to stick his or her neck out.
There should be an outcry from the ad community as loud as what we’re hearing about viewability. Why continue to waste money on ads that can’t be seen? And why continue to throw away 99.9% of your money on ads you say aren’t doing anything?
The difference is that those 99.9% of ads are working – and also that unlike viewability, where metrics are still a question, attribution is a known quantity where the only thing that needs adjusting is the formula. You can easily find out who’s seen or clicked on which ad, when it was served, and who’s doing the serving.
It’s only a matter of figuring out how to attribute and how to prioritize.
And while you’re at it, locking in your attribution formula a whole year in advance won’t cut it anymore, either!
Brian is a proven expert and leader in revenue optimization and business analytics. On behalf of search, media and startup companies, he has achieved substantial success in growth and led transformative analytics, revenue, and new business teams that frequently bested company goals. A driven and strategic leader well known for his sophisticated analyses, Brian most recently spent three years with Twelvefold Media where he served as VP of Revenue Optimization. Prior to Twelvefold, he advised tech start-ups, including Media River, Tapatap, Clickfacts and KPG Ventures, on revenue strategies, team development processes, and sales operations. Earlier in his career, he served seven years in various positions at search media company LookSmart, including the most recent title Vice President of Business Analytics, where his critical analysis and team building were instrumental in driving the financial performance of the company and besting aggressive margin goals.