Game Theory, part of economics that deals with the interaction of competing behaviours, and PPC have a lot more in common than you might think. The entire idea of PPC being more like a bridge between marketing and IT isn’t true in the slightest, since the theory of behaviour (which, granted, a lot of marketers rely on) and auctions (bidding) are found within economic theory.
Game theory in particular has a lot in common with the practise of PPC because of the sort of interactions involved. A PPC campaign is created by an individual who has expectations of a searcher’s behaviour, but the model by which those ideas engage with the searcher is very much up to the network and the publisher.
A typical example is when to bid more for certain terms and when to bid less. The PPC expert will examine the term, the conversions, the AOV, the volume of searches, time of day and variations of search term in order to find the optimum bid to create high volumes of traffic at a decent volume. However this idea competes with the user expectations, since users don’t act in the way that economists would like. In economics you expect self interest, however in the pool of results on a search page there’s not a great deal that indicates which would be of the most benefit and semantics within ad copy cannot be converted into a bidding metric with great ease.
Here’s an example of competing arguments within ads.
A user is searching for clothing and types in a generic term. The system itself doesn’t have a lot of information since users don’t typically type in their budget or more details about what they want. A PPC expert would have an ad appearing designed to get the most clickthroughs possible, relying on the historic conversion rate of the term, cost per click, ROI, basket value (dependent on how well the site upsells), etc.
The user clicks and sees that the average price is maybe £20 for their product so they bounce off and that’s a wasted click. It’s no ones fault because the user shouldn’t have to search for budget terms (even “cheap” as a term doesn’t mean much because it’s a value term not a price term most of the time).
With the “model”, it seems like this is great because you’re still making margin on the other visits who do convert at a £40 basket value, for example, so the model says this is the equilibrium and that it should remain. The PPC expert doesn’t touch it because why would you risk good margin and volume?
But the canny professional will test the game of searcher and site by testing some ads, something that can’t be broken down within the model. If you say in the ad that the item of clothing starts at £20 for example, you no longer have people clicking and being put off by the £20 price because they are aware of the price on the site. So immediately you are having an impression share for your generic term but avoiding clicks from the wrong users which is something too few professionals think about.
The model is flawed because it doesn’t know the reasons behind bounce and abandonment – it’s breaking down success into metrics and ignoring the economics. The economics state that there are infinite number of “markets” for each bid because while people are searching they have expectations that aren’t present in the query – such as what they’re willing to spend, if they’ll actually convert, if they will be a valuable customer, etc.
If you’re a PPC amateur, or an economics nut, then I suggest looking at game theory and PPC together because it’s really very interesting.