Optimize conversions in ads for mobile apps

The strategy allows you to get the maximum number of clicks resulting not only in app installs, but also in targeted actions inside the app. This strategy is good for advertisers who know their specific goal for attracting new users to their app.

How it works

You set an action to be made by the user in the app as a goal, for example, adding an item to the shopping cart or signing up. This strategy aims to increase the share of clicks on ads that most often produce conversions. Bids are automatically selected in such a way as to achieve the best result using all the specified criteria.

Ads will not be served to users if it is determined that the app is already installed on their phone or tablet.

Learn more about how this strategy works.

Prerequisites for using the strategy

The app must be connected to one of the mobile app attribution platforms supported by Yandex.Direct. A tracking link must be specified for your app in Yandex.Direct.

Event reporting must be enabled in the mobile app attribution platform. Events must be passed with the names corresponding to your Yandex.Direct goals.
Goals in Yandex.Direct
  • “App installs”.
  • “Goal achieved”.
  • “Launched app”.
  • “Added payment info”.
  • “Added to cart”.
  • “Added to wish list”.
  • “Completed registration”.
  • “Completed tutorial”.
  • “Initiated order”.
  • “Made a purchase”.
  • “Submitted rating”.
  • “Performed search”.
  • “Spent credits”.
  • “Achievement unlocked”.
  • “Content viewed”.
  • “Time spent in the app”.
  • “User shared”.
  • “Custom event 1-3” — use this name if other names do not fit the event.

Read the instructions in your mobile app attribution platform on how to set up and pass events.

Strategy settings

Goal

A goal is an action you expect from a user who was exposed to your ad. Select the event passed by the mobile app attribution platform as the goal. App installed is selected by default.

Average CPA

The ratio of weekly spending to the number of conversions.

During the first week after activation, the strategy is trained and the actual value of this parameter may differ from the set value. After this, you might see deviations in the average cost per action (either up or down) over the course of a week, but by the end of each week the value will get closer to the set value.

The minimum value for average CPA is 0.03 TL.

Weekly budget

The minimum weekly budget is 17 TL.

The actual expenditure of a campaign using this strategy may vary by several percent from the specified budget. Accuracy may be affected when you switch between strategies, make changes to the budget, suspend a campaign, refine your time targeting or change the campaign start date.

Budget distribution depends on your time targeting settings. For example, if you limit ad impressions to two days a week, your budget will only be distributed over those two days.

If the budget cannot be spent, you will see an alert in the interface and get a notification sent to your inbox. This usually means that the keywords selected have a low CTR, low budget or low CPC.

Maximum CPC

You can limit your CPC. If the maximum CPC is not set, by default it will not exceed 10% of the weekly budget.

Note. Set relevant parameter values. If your average CPA is unattainable, budget is insufficient, or maximum CPC is low, the system may not be able to raise bids for effective keywords, and the number of conversions will decrease.

Pay per installation

The strategy lets you get conversions at a fixed price per app install.

Strategy settings

Enable the option Pay per installation. The goal will automatically change to "App installs".

Enter CPI.

For the strategy to be effective, transmit installation data from all sources to Yandex.Direct. If you use AppMetrica, you don't need any additional tracker settings. If you use Adjust, enable the Yandex.Direct module. For AppsFlyer, enable the option to send installs attributed to any partner or organic.

Specifics of CPI campaigns

When a campaign has few or no conversions, a warning appears in the interface.

How to calculate the optimal conversion cost

The optimal conversion cost is the value that provides the maximum total profit.

Profit per customer can be calculated using the formula:

Profit = Conversion value — Conversion cost

To estimate the total profit, multiply this difference by the number of conversions:

Profit = (Conversion value – Conversion cost) × Conversion count

The conversion value is the profit to the business from the conversion. For example, if 100 requests brings you 50,000 rubles, the conversion value for the goal “Submitted an online request” would be 500 rubles.

The selected conversion cost affects both profit per customer and the number of converted sessions. When the conversion cost is low, profit per customer may be high while the number of converted sessions may be low. On the contrary, if the conversion cost is high, it may reduce profit per customer but lead to a larger number of conversions. Find a value for the conversion cost that would maximize your total profit.

The number of converted sessions does not have a linear relationship to the the conversion cost. Once you reach a certain threshold, making the conversion cost higher does not significantly increase the number of converted sessions:

Use Report Wizard to find out the current conversion cost by selecting Conversion rate (%), Goal cost, and Conversions. You can calculate the optimal conversion cost through an experiment. Specify different conversion cost values in experimental campaign settings. The experiment must run long enough to accumulate sufficient statistics for proper result analysis (at least a week). Use the statistics to estimate how the change in the conversion cost affected your profit.

When comparing the statistics, consider the statistical error, i.e., possible deviation of the data obtained from the true value. The table shows the relative error depending on the number of conversions.

Conversions Deviation from true value
Conversion rate (%) CPA
25 ± 20% ± 40%
100 ± 10% ± 20%
400 ± 5% ± 10%
2500 ± 2% ± 4%
10,000 ± 1% ± 2%

Learn more about how to create a report and make decisions based on statistics