Users do not always make a decision to buy a product or service immediately after seeing an ad. Even if they are interested in the offer, they can make a purchase later by entering a query in a search engine or following a link from an advertisement on another site. Therefore, when analyzing conversions, it is important to take into account not only the last interaction, but also the influence of the advertising channel on the decision to purchase a product or service.
Goals and hypotheses of the advertising campaign
As part of an advertising campaign in myTarget to promote a mobile application on iOS and Android, experts from the BlaBlaCar road trip marketplace measured the number of post-view bookings (made after viewing ads). To do this, the team conducted a special test on the platform for three months.
During the campaign, advertisers planned to analyze how many reservations made after a click on an ad (post-click) would have to be booked after an ad was shown, but without a click on it (post-view). The total cost of attracting post-view bookings (CPA) should not exceed 29 rubles.
For the test, all regions of Russia in which advertising was launched were divided into two groups: test and control. At the same time, the audience of the two groups was the same in terms of the number of bookings and their frequency over the past 30 days.
Campaigns were launched for test regions in myTarget, and control regions were excluded from impressions on the platform. Thus, the only difference between the first group and the second was in the work of the myTarget advertising campaign.
When setting up ads in test regions, 25% of the budget fell on a square video, 75% on a multi-format and a banner. At the same time, 21% was spent on advertising campaigns with a pay-per-click (CPC) model, 72% – with optimized pay-per-1000 impressions (oCPM), 7% – with pay-per-install (CPI). We chose this approach to budget allocation, as it has already shown high efficiency earlier.
The targeting was for men and women between the ages of 20 and 55 who were interested in travel. The remarketing tool also made it possible to serve ads to users who searched for trips but did not click on the ad, as well as to those who studied the service offers but did not book trips.
Example of BlaBlaCar ads
In three weeks, active campaigns launched on myTarget generated 6.4 thousand post-click bookings. When comparing the control and test groups of regions, it turned out that the test region showed 24 thousand more bookings than the control one. Thus, for 1 post-click booking, 2.7 post-view bookings were received.
These results were obtained within 7 days after the start of the test. This may be due to the fact that the user did not start the trip immediately after seeing the advertisement. The total cost of attracting post-view bookings was 22 rubles, which is 25% lower than the KPI.
Schedule reservations after viewing advertisements in myTarget
The obtained results show that within the framework of performance campaigns it is not always worth analyzing only direct conversions; it is also important to take into account the delayed effect of promotion.