Every comparison engine in the world basically uses two ways to invoice their merchants; either on a CPC (cost-per-click) basis or on a CPS (cost-per-sale) basis. I have talked about the advantages (and disadvantages) before but I will briefly explain that again:
- CPC: favored by comparison engines because it is transparent and it can be measured by both parties. CPC however poses more risks for the merchant, as they are uncertain on how the visitors will convert on their website. But, CPC puts the risk where it should be.
- CPS: favored by merchants, as their risk is zero. If visitors convert, then they have to pay the comparison engine. The comparison engines are dependent on the merchant for conversion numbers. This is quite a tricky model as comparison engines cannot be held responsible for ‘crappy’ navigation on the website of the merchant, wrong links in datafeeds, servers that are down, etc. CPS thus shifts a lot of the risk to the comparison engine, but they are not capable of mitigating or handling that risk. Therefore we believe this model is not the one to go for as a comparison engine.
Comparison engines normally charge merchants a certain fixed amount per click per (sub)category. There are also some comparison engines who use that model, but with a twist. For example in The Netherlands Kieskeurig.nl uses a CPC pricing, however the level of the CPC depends on the price of a product. So, the higher the price of the product, the higher the CPC. This looks like a fine idea, but the higher the price of a product does not necessarily mean the margin on that product is higher as well. Some products like lenses (you know, to put in your eyes) or ink cartridges have extremely high margins. In the model I just mentioned the CPC’s (the one from Kieskeurig) for those products would not be accurate. If you understand the dynamics behind comparison engines, this could mean that they will be outplayed by competitors who do charge an accurate CPC. These competitors can spend more money on promoting these categories, doing marketing campaigns, etc.
For a long time we (as Compare Group) have been thinking about a good and fair way (both for shops as ourselves) to determine the CPC levels. A lot of articles have been written about this subject and the clear message is that the current way of CPC pricing is one designed for the short term and that comparison engines should make themselves ready for the next era.
Quote taken from an article: “Are You a Dumb Grasshopper?”
“Dumb advertisers will eventually learn to be smart. Dumb CSEs can fix their business models today and salvage their relationships with their advertisers, or they continue to rake in some extra bucks today at the expense of tomorrow.
Back when I was a dumb kid, I seem to remember a tale about a grasshopper who lived for today and an ant that prepared for tomorrow. When winter came, the grasshopper had no food and died.”
I recommend you to read some other articles on this subject, like “Comparison Engines Losing Steam? Shoppers Balk at CPC Model” and “Shopping Engines in the Dark Ages“.
The Compare Group has taken this serious and has been working on a way to tackle this problem. In Q4 of this year we will introduce a model that takes several dynamics into account when calculating the CPC price (basically just like Google calculates the price of their Adwords). This model for example will charge lower CPC’s if the visitor demand is low and it will also charge lower CPC’s if the demand in products is low (so, adding products in long tail subcategories will be cheap). On the other hand it will charge higher CPC’s for categories that are crowded by merchants and visitors.
We still have to see if this will work out fine, but at least we are trying. And, we really hope that merchants will see us as their partner to boost their sales and not just as a supplier of clicks. We are really fed up with the discussions about CPC… To add to that, I have never seen merchants complain about paying extremely high CPC’s for their Adword campaigns, so I trust that this will also be the case for our new pricing model. Hey, it’s not us determining the CPC, it’s the model ;-)