When an advertiser signs up for a paid search campaign, he expects to get a certain type of traffic. The traffic should consist of users who actively go to a search engine, type in a keyword the advertiser has specifically targeted, then clicks on the ad to visit the advertiser website. There are a variety of ways that search engines and shady affiliates have subverted this process. In this post we’ll review one way this expectation is being subverted by the Bing/Yahoo search network.

The Telltale Signs

The key indicator that tipped me off was when two nearly identical keywords showed dramatically different click through rate. The average click through rate for my phrase match keywords on Bing was 4-6% for months. In July of 2018 suddenly one of my keywords went from getting 50-200 impressions per day to 5-10k impressions per day. At the same time, clicks were basically flat. Here’s a screen capture of the performance data for two of these keywords

The Example

At this point, I’m wondering, what’s the explanation? The most straightforward and honest explanation would be that suddenly people are searching one keyword tens of thousands of times per day more often than they were last week, but only that exact keyword, other synonyms to it were performing as expected. Seems unlikely right? I couldn’t come up with an explanation that sounded reasonable, so I just lowered the bid for the keyword which was performing strangely and went back to looking at cat pictures on Facebook. That’s when I found it.

Oh, so Yahoo is buying ads on Facebook. “Checking Account Offers” just happens to be a keyword with many advertisers competing and a very high cost per click, so what happens when you click that ad?

Oh, would you look at that? It’s a search results page that only shows ads, (well if you scroll down far enough you’ll find some organic results, but nobody does that) how weird. This is a very old game called arbitrage. Basically Yahoo knows it can buy a click from Facebook for say $1 or $2. They also know that if they make that click look like it was a search for “checking account offers” they could rake in upwards of $10 from a single click on that search results page.

Now it makes sense

The global average click through rate for legitimate non-branded paid search is about 2-4%. The global average click through rate for display ads is about .05%-.2%. Looking at the first screen capture showing the performance data for the two synonymous keywords we can clearly see that the first keyword is bacically display traffic (because Bing/Yahoo are using this arbitrage method to drive it) thus the .09% CTR and the second keyword is actual search traffic thus the 3.89% CTR.

Why should I care? Isn’t traffic just traffic?

The reason I care is that someone scrolling through cat pictures on facebook who clicks an ad has an entirely different intent than the user who actively goes to a search engine and types in a keyword. This is reflected in the click through rate on the front end, but also the conversion rate. Whether your conversion is a completed lead form or online purchase you’ll see significantly different performance between the search and display channels.

So what to do about it?

I looked into what my options might be. At first I tried pausing the specific keyword that was driving the low intent/arbitrage traffic. The problem was this just became a game of whack a mole with the engine simply changing the target keyword from the one I paused to the synonym which was still active. I was already excluding search partners/network, so that wasn’t an option that would resolve the issue. The solution I arrived at was to simply stop buying Bing/Yahoo search traffic for all non branded terms.

While others might arrive at a different conclusion, as an advertiser who only cares about direct response performance I just had to pull my spend. Paid search provides an acceptable level of performance for my business goals, which is why I spend money on it. Display does not. Since Bing/Yahoo decided to abuse my ad dollars by selling me display traffic as if it was search traffic in order to capture more of my budget, my only solution was to walk away. I hope they decide to reverse course on this sometime in the future, but given their small and declining portion of genuine search traffic I’d expect them to continue using these shady tactics until they finally give up on the search ad business altogether.