As you are by now hopefully aware, on February 13th 2013 Google are launching one of the biggest ever changes to the search landscape. That is of course, the changes to take Google Product Search from its current 'free' model, to a paid CPC model and blending it with the existing Product List Ads (PLAs) in Adwords.
If you're sitting there scratching your head, thinking this is the first you've heard of the changes, where have you been hiding?! Fear not though as we've got you covered, take a look at our white paper which explains everything you need to know.
The first in this subsequent series of articles started to look at the complexities of Bid Strategy and exactly how the 'Grouping' that Adwords allows will affect your costs. We took a 'flat-bid' structure that would apply to Top Sellers and in short, the stats showed that grouping products in this way simply does not work effectively for optimisation, if for no other reason than that every product has a unique value to your business and therefore demands being treated as such.
We're going to look to build on that article now by taking what is arguably a more intelligent and more complex product grouping and actually using a tiered margin model to base bids on; something which has been recommended by Google.
Using the historical data we have for our merchants, we're able to comparatively model out bids and performance at any level of grouping and offer recommendations around which would be most effective.
(For the purpose of this article, the CoS refers to the function of cost divided by margin value, normally sales revenue)
Scenario 1: Top 50 products grouped based on a high/average/low margin tiering system.
For the merchant used in this example, we have again started with the Top 50 products over a given period of time. These products have then been segregated into 3 tiers based on their actual margin values; High, Average & Low Margin. The total margin value this group of products represents is in excess of £125,000 and varies anywhere from £50 to over £300 per product.
If we were to assume the margin groups to be as follows:
High Margin = £200< with a bid of £0.60
Average Margin = £100 to £200 with a bid of £0.40
Low Margin =
Using the data we have available, we can see that this would have resulted in total costs of £11,109 and a blended CoS of 8.85%, which is almost 4% higher than the target CoS that we are working towards, at 5%.
However, this is not nearly as poor as when compared to a completely flat bid structure (applied in the same way as we did in the first article) as the CoS would have been even greater at over 11%, with total costs coming in just shy of £14,000!
Scenario 2: Every product bid individually calculating an effective CoS based on actual margin value
With the automated systems that we have developed, we can actually build a single PLA for each of the individual products, bidding these independently of each other, exactly as you would on CSEs such as Shopzilla (though with free reign on bids, starting anywhere from 1p!). Based on a granular product level structure, we can easily calculate the starting bids that these products should enter the Google Shopping space with.
Taking the same 50 products used in the first scenario, we can work back from the 5% CoS target that we have in place for the merchant to calculate the bids needed in order to deliver efficient sales. In doing so, the bids for these products now range anywhere from 8p right up to £1.83. The total costs in this instance would be £6,275, which is some 43% lower than in the tiered model we looked at in the scenario above and 55% lower than in a completely flat bid structure.
And this is just based on a subset of 50 products that are considered the best performing...
To answer the question posed in the title, margin optimisation is ultimately what every retailer should be striving for, but, it has to be used at a product level to be effective. Ignoring product level margins and optimising based on tiers or groups of products will render the information useless and ultimately, defeats the point of acquiring product level margins in the first place.
Article written by Mark Batson, Client Performance Manager