Full-Funnel Attribution: All in the Details

1
582

detail_smallADOTAS – A little more than 10 years ago, I had a meeting I’ll never forget for the rest of my life.

It was me, my boss and my boss’s boss, a C-Level exec at our NYSE-traded company. The pace of growth was insane — I was hiring people left and right, and the three of us met to review plans.

I took a spreadsheet into the meeting prepared by one of my managers. With little time left before the meeting, I did not review the manager’s work preparing the spreadsheet–I simply trusted.

Once past the first page, the C-Level exec spotted a few errors. It was embarrassing. Then he spotted more errors. And then it happened.

He put everything down and looked at us. “Do you know what business is about?”

We sat speechless.

He continued, “Business is about details.” And then he walked out of the room. The meeting was over — he left me and my boss holding mistake-ridden spreadsheets.

That was one of the most important lessons I’ve ever learned.

This year you may dive into one of the most eye-opening areas of online marketing — full funnel attribution. If you’re managing a multimillion dollar online ad budget, you’ll learn an important business detail: today’s outdated ad-tracking systems give 100% transaction credit to the very last clicked or last viewed ad before an online transaction.

Example: if four Internet ads contribute to a transaction, today’s outdated systems allocate entire credit to the fourth and last ad, ignoring the first three ads that actually drove the revenue. Zero credit to revenue drivers, and 100% credit to the last ad placed. Frightening.

But now you can begin evaluating full-funnel media attribution. Full-funnel attribution systems recognize that credit should be assigned to a team of Internet ads versus the last ad.

Seven-figure efficiency in your online ad budget? Fantastic! But before you foray into full-funnel attribution, ask seven questions because they reveal key details.

1. Will the attribution system work with every form of paid media?

Quite often, you’ll discover attribution systems can’t handle certain affiliate programs, or they only work with search, but not display, not email, etc. They should either work with every form of paid media (plus organic) or they don’t. If an attribution system doesn’t embrace everything… it’s not full funnel.

2. Are there any time limits for tracking and attributing?

Many of the leading ad servers with attribution modeling bolted on have time limits. Meaning after xx days, nothing else is tracked. Nothing.

When did consumers decide they were going to buy a product or service on an ad server’s arbitrary timeline vs. their own timeline? When you track and attribute without limits, you may be surprised by what you find.

3. Does the attribution system cap touchpoints?

After asking this question very carefully, you might discover that well-known systems purporting attribution, cap out at 10 events. Yes, after 10 events (view-throughs or click-throughs) certain systems just delete data — keeping only the most recent 10 views and clicks. And if you delete that valuable information in the upper funnel… revealing the media sources originating or driving your revenue –that’s nowhere near full-funnel attribution.

4. Is view-through attribution done in real-time?

Are view-throughs incorporated into the attribution system in real-time or uploaded after-the-fact? With explosive growth in real-time bidding, those real-time bidding systems need real-time feedback. Make sure your attribution system incorporates impression-level data into real-time pixel firing.

5. Is there selective pixel firing?

Is the attribution system using intelligent technology for pixel firing? Meaning, if you pay a high bounty for new customers, but a lower one for return customers — can your attribution system tell the difference? If you define a return customer by number of days from last transaction and change that definition, can your system synch in real-time and selectively pixel-fire based on your definition-set?

6. Is the attribution algorithm transparent and explainable to your CEO?

Is the attribution algorithm transparent, or is it a black-box formula with an equation longer than your arm? Can you run dynamic yet simple “what if” scenarios in a data sandbox where historical data never changes–only the assigned attributions? Can you outline your attribution algorithm for your media partners and CEO in 30 seconds so they understand it?

7. How much time do you waste translating data into knowledge?

Like most marketers, we’re drowning in data but thirsting for knowledge. Are there tons of fancy graphs, and weighty outputs taking weeks to analyze? Or does the attribution system crunch all the data in the background, serving you an elegant, single metric for fast decision-making.

So before you foray into full-funnel attribution, ask these seven important questions. Because your success in business this year… is about details.

1 COMMENT

  1. Finally! Someone else realizes that last click attribution creates misleading data that is worse than no data. Some track first click but that is no better.

    What you must do is track ALL clicks. I said misleading but totally wrong might be more accurate – and I can prove it.

    Select a product on an ecommerce site that has been offered a long time but shows no sales. Create AdWords ads for that product. After you have at least four sales notice what got credit for those sales.

    Here is what I have seen repeatedly in Google Analytics. One sale will be credited to the ads; one to organic or direct and two to other sources such as Google Base or Shopping Comparison sites.

    Why that is obviously wrong: the product page was already listed in organic search and was already in Google Product Search and on those shopping comparison sites – sometimes for years – and never generated a single sale before.

    If there has NEVER been a sale of that product until you created ppc ads and then you sell four the sales were CLEARLY generated by those ads and NOT anything else.

    Businesses are making important decisions based on data that is dangerously misleading and they don’t even know that is what they’re doing.

    Bad data = bad decisions = business failure.

    It took me far too long to find out about last click tracking – and I only did because I knew who to ask and kept contacting Web Analytics experts until they finally answered to my satisfaction.

    It is about time that this analytics people make it clear that conversion tracking is not accurate and make improving it their priority.

LEAVE A REPLY

Please enter your comment!
Please enter your name here