Do retailers need a complicated PPC bidding strategy? Yes. Undoubtedly, yes.
Managing your bids by simply shifting based on a simple ROI model is better than nothing. But it’s not the most efficient way of getting the most out of your keywords. There are many other things to consider.
Firstly, you need to understand the margin of the products being purchased – and you must have as much detail as possible. Your overall target could be 5% cost of sale, but if your margin is only 2% on technology products then these terms should be managed to a different number.
At Summit, we split optimisation as much as possible, depending on the level of margin data our clients can provide.
Another thing to consider is the products purchased against the keywords used. A user purchasing on the term ‘PlayStation 3’ does not mean that user purchased a PlayStation 3. For some clients, we’ve seen this cross-over of keyword to product as high as 60%. So reports should be pulled regularly to analyse this cross-over, otherwise any optimisation based on margin could be way off the mark.
Bid gap is another key point to understand. Each position for each keyword within each separate auction will have varying bid gaps due to Google’s quality score system. But there are models that can be built to predict these based on previous results. For example, if you know the estimated bid gap for a keyword is £0.10, then a positive performance leading to an increase of £0.02 may not be enough to push your AdText up a position.
On a similar note, performance by position should be taken into account – and this is especially relevant for keywords in the top three positions.
Regularly review whether a keyword gets more traffic (or a different conversion) in position 1-3. If not, why be in position 1 – it’s just burning money for the sake of it.
The key strategy here is to outline all of the inputs and outputs that can influence the performance of your PPC campaigns: do not make a change based purely on the sale that each keyword is currently driving.