Tests - of any type - cost money. Over time, you run the risk of wasting money on a strategy that doesn't work or doesn't work as well as another strategy. On the other hand, you Industry Email List need to run your test long enough to get a clear answer as to which test group performed best. A 50/50 test split means cutting your total traffic in half. You want to make sure that the variations receive statistically sufficient traffic within the time frame for the test. How long it takes will depend on how much traffic you get and how obvious the difference is. Keep in Industry Email List mind that any tool (or agency) that uses algorithms and machine learning to manage auctions will need some time to collect data before the automation really lives up to its potential. .
In this case, the retailer ran the test for 10 weeks. You can see that the differences between us and their existing tool were very slight in the first half of the Industry Email List test as the algorithms collected data. But by the second half of the test, we had collected Industry Email List enough data for autobidding and the new campaign structure to take effect, creating a much more pronounced difference between our two approaches. Another important thing to remember is to wait another two weeks after the end of the test before collecting and evaluating the results in order to capture latent sales. Assess the results Before the test, be sure to set yourself clear and achievable goals.
The goals you want your candidates to achieve should be achievable for the parameters (e.g. budget, time, season, product bias) you set alongside them. You Industry Email List can't expect the moon just yet. Now is not the time to get fancy with your metrics either. Set the same type of goals you used so you can easily compare new results to your baseline data. Once your test period is over, you can compare the results of each test / agency with your key performance indicators. In Industry Email List this case, the most important metric for the retailer was who could generate the most revenue within target ROAS. However, they also looked at other metrics such as cost and ROAS. You're not necessarily looking for