We have often been asked how our publishers can optimize for Quality Click Pricing and what data they should use to do so. Although we don’t disclose the QCP algorithm, the majority of elements that go into it are available in the reports that are already available to you in eBay Partner Network. This article aims to show you the different techniques and ways of manipulating the data to help you to get a clearer idea of how you can optimize your traffic and thus increase your earnings. The importance of campaign ids
As your EPC is calculated at campaign level, it is very important that you split your activity down into different campaigns, so you can see the EPC for each segment of traffic that you are sending to eBay and optimize accordingly.
There are a number of different ways to do this: • By site if you have many of them • By placement • By tool • By category • By keyword group, if you are buying paid search traffic
Publishers who use paid search to drive traffic to their sites have found this very useful, because they can easily compare what they are paying a search engine for a group of keywords to the EPC they are earning.
We recommend that you split up your activity into campaigns as soon as possible and then don’t change them too often, as when you start a new campaign you may see a slight change in your overall EPC for a few days while you build up a history. For the same reason, if you are new to eBay Partner Network, it is best to set up your campaigns from the very beginning to avoid having to do it later. Also, ensure that your campaigns are not too small – to get the most accurate data, try and keep the click volume above 100 clicks per day. See question 29 of the FAQs for more information on low volume campaigns.
Using the reports
We have many different reports available that show your performance in slightly different ways. For a summary, it is good to look at the By Day report, as it shows your EPC at by program. In order to get a deeper understanding of how your traffic is performing, you should look at the campaign level EPC.
Although EPC is the most important metric for you to optimise for, the following metrics may also be useful: • Conversion Rate from click to sale (calculated by dividing number of winning bids by number of clicks) • Conversion Rate from click to ACRU (calculated by dividing number of ACRUs by number of clicks) • Revenue per click (calculated by dividing winning bid revenue by number of clicks)
Although it is sensible to keep an eye on your daily EPC, in order to get the most accurate idea of your EPC, we recommend that you look at the data over a longer period before making any conclusions about changes you need to make. This is because small volume publishers especially will see fluctuations day to day, and if you are a new publisher, it will take time to establish historical data.
The QCP algorithm is designed to reward affiliates who get people interested in shopping on eBay, whether that is within a specific category related to an affiliate’s site or more generally. Therefore, if you have a niche content site, as opposed to a more general shopping site, the category in which the users you send to eBay make purchases is a good indicator of the quality of your clicks. For example, if your site about fishing mainly converts in the fishing category on eBay, your site is likely to be highly targeted. On the other hand, if the same fishing site mainly has sales that convert in the book, DVD and clothing categories, you may want to do some further optimisation. This is important because the time between click and purchase is an important element of the QCP system and the more targeted the traffic, the shorter the time between click and purchase is likely to be.
To see the categories from which your users buy, you can use the Category report, which you can find in the interface. You can also view the data at campaign level, which can be useful for publishers with many niche sites or who have one site with many categories, as long as they are broken down into different campaigns. You can view the data split by a number of different metrics, but as the main component of the algorithm is still winning bid revenue, this is perhaps the most useful way of viewing the graphs.
For more information on the category report go to the user manual, select “Reports”, then “Summary Reporting”, then “By Category Summary.”
More Advanced Analysis – Transaction Download Report
If you want to know as much as possible about your traffic’s performance, then you should take the time to really understand what is in the Transaction Download report. The report gives you details about every single event that happened as a result of your traffic, including CRUs, bids, BINs, winning bids/BINs and winning bid revenue. If you wish, you can also pass through a custom id within your tracking link and this will appear in the report. This could be particularly useful if you drive traffic to your site from paid search and want to pass through an id that links to the KW that you bought. Using custom IDs can also be helpful for optimizing campaigns that do not have enough traffic to receive an individual EPC.
Perhaps the most important information in the report, at least in relation to EPC optimisation, is the click timestamp and the event date, especially when used in conjunction with the winning bid revenue amount. Under QCP, we value sales that happen more closely to the click more highly than those that occur towards the end of the 7 day cookie lifetime. If you look at the click timestamp and the event date, you can tell to within 24 hours how long the sale took to convert after the click took place. If you then sort the report by campaign or by custom id, you can then calculate the amount of winning bid revenue that occurred in the first 24 hours after the click, per campaign or custom id. This, combined with the daily EPC data in the By Campaign report, is powerful data with which to optimize. To take this even further, you could then carry out the same exercise, but use the meta or leaf category id to cut the data instead, which would show you which categories yield the highest revenue in the first 24 hours.
To find the Transaction Download report, go to the Reports tab, then select Transaction Download report and highlight the Download All Events button. For more information visit the user manual, select Report and then Transaction Download.
Once you have evaluated your campaigns and have identified areas that you want to improve, you are ready to make changes. We recommend that after any change you make, you wait a period of time for the EPC to stabilize before making any conclusions on its effectiveness. With completely new approaches, it is probably best to set up a new campaign rather than using an existing id, and start by sending only a small amount of volume (but not under 100 clicks!) That way, if the new activity does not perform well, then you have not diluted your overall program EPC too much.
If you have the capability to do AB testing, then you should definitely utilize this when making changes. If you don’t, then a very good alterative if you have more than one site or many categories within your site, is to make the change in one site or category first and if it results in an increased EPC, then roll it across the rest of your portfolio. If this isn’t possible then try pre and post testing – in other words, look at your EPC before and after you made the change to see the difference. When using either by site, by category, or pre and post testing, remember to adjust your results for any external effects that may impact your performance – for example, seasonal changes such as the ramp up to the Holiday season. Also, try to make sure that you have not changed anything else on your site, including your sources of traffic, as these changes could skew results.
These techniques are some that we have used when helping publishers optimize on a one-on-one basis. However, with the number of publishers that we have, many of whom who are earning considerably more under QCP than they were under the old CPA model, we are sure that there are many other ways that you have found to analyze the data to help you improve your EPC. And we would love to hear them, so if you have any words of wisdom for your fellow publishers, please comment below!