Set target KPIs by device type, as mobile and desktop have different performance indicators in different products. This is especially important when working with tCPA in Google Ads and the CPA model in Yandex.Direct.
In Yandex, use only mobile ads for mobile devices updated 2024 mobile phone number data and only desktop ads for desktop devices.
For the B2B sector, we recommend keeping an equal share of traffic. For the B2C sector, we recommend splitting towards mobile in a 60/40 ratio.
Analyze the "reach" of users at each step when using a long product application funnel.
Geo: CR and CPA
The traffic volume in the regions is higher than in Moscow and St. Petersburg. Geotargeting depends on the location of the office. High CR and low CPL in the regions is explained by low competition and ease of optimization.
Moscow has high competition in the B2B sector, and, as a result, low CR and high CPL relative to the regions. Since the LTV from the client is high, Moscow is important for promotion.
Saint Petersburg is the least efficient region. The competition for search is similar to Moscow, but the purchasing power of the population is lower.
Search advertising of financial products: Geo CR and CPA - how to optimize CR and CPA taking into account geography
How to optimize CR and CPA taking into account geography:
Use CRM data: statistics by city across the entire funnel to determine AR,% and SR,% of the city.
Divide regions into separate campaigns if they account for more than 60% of the total budget. This will allow you to split the budget based on the CR, AR, and SR of the city. Example of division: high-, medium-, and low-conversion cities. If KPIs are difficult to achieve, we recommend narrowing geotargeting to the most effective cities.
Assign different tCPA for automatic strategies or optimization on any section, since the approach and conversion rate for geo-segments are different.
Use the most effective budget split. According to our data, Moscow - 20%, St. Petersburg - 15%. High-conversion cities - 20%, medium-conversion - 40%, low-conversion - 5%.