Direct
Help
Contact us

Automatic management

Using the automatic bid placement strategy, the system manages placement independently, aiming for as many targeted click-throughs as possible. You decide what your primary goal is so that the system can orient itself around it, and set expenditure limits. The system then undertakes all necessary calculations and allocates the budget so that funds are distributed in the most economical possible way.

Automatic strategies completely solve bid managment issues, allowing you to focus on more important indicators. For example, you can focus on the quality of ad texts, choice of website landing pages, or on calculating and analyzing the revenue you receive from the users your ads attract.

Average CPC

This strategy ensures the maximum number of clicks for the average amount you are prepared to pay per visitor.

Average cost per action

This strategy focuses on maximizing the number of converted sessions to your site and makes sure that the average cost per such visit for the week doesn't go beyond the value you set.

Average Cost to Install App

This strategy is only available for ads for mobile apps.

This strategy allows you to get the most possible app installations for an average weekly price per installation that comes as close as possible to the value you've set.

Average return on investment

This strategy allows you to maximize conversion for the goal set in Yandex.Metrica and bring your goal's average ROI for the week in line with the value you set.

Weekly budget

The set weekly budget is distributed in order to achieve the result you are aiming for, whether it be, for example, as many click-throughs to your site or app installations as possible

Weekly click package

This strategy ensures that you get a certain number of click-throughs to your site or vCard every week at a minimum CPC.

How automatic stragies work

All Direct automated strategies are based on a special mathematical algorithm that forecasts how likely one is to click-though from your ad and reach certain targeted goals on your site.

When generating a forecast, the system analyzes the statistics by keywords selected for the ad. For search sites, the system takes impression data from Yandex search into account. For content sites, it uses impression data from ad networks (YAN and ad exchanges). If there aren't sufficient statistics, it uses a database of analogies and information about click-throughs from Yandex search results. The system also takes the behavior of users who click on ads into account. The most effective phrases are chosen, i.e. those that draw the most clients to your site, and ads are displayed more frequently based on those phrases.

Forecasts are formulated by the hour for every day of the week. This way, the relation between the number of impressions, clicks, and targeted events and the time of day/day of week can be determined. The forecast is regularly re-calculated using the latest data during the placement process.

In order to manage bids as effectively as possible, the automatic strategies make adjustments to ad campaigns far more frequently than you could do manually or with the help of third-party ad management tools. Bids are loaded into the Direct interface every 15 minutes. Automatic strategies can change bids more often than these changes are displayed in the interface.

Protection from over-spending is built into the automated strategies by default. The system cannot spend more than 35% of the weekly budget per day, and the bid never rises above 10% of the weekly budget or 450 руб. (the limit for other currencies will be recalculated at the current exchange rate of that currency against the ruble).

How to use automatic strategies effectively

General recommendations

For automatic strategies to work correctly, it is very important that your ads' redirect pages be available most of the time, and that your ad campaign contain enough ads and keywords. Ideally, the campaign should advertise one website (or mobile app) and have stable indicators over the course of a month.

To achieve the maximum results, the system conducts a detailed set of calculations that serve to regularly optimize your settings. For this reason, it's important to give the system some time to work everything out. In most cases this takes a few weeks.

If you primarily use infrequent search terms, then the system will need more time to forecast click and conversion figures as well as to determine optimal spending. Expand your keyword list to help the system.

Determine your settings before the launch so that you won't need to alter them while your strategy is running. It will be more difficult for the system to formulate a quality forecast if you frequently re-launch your strategy or temporarily switch impressions over to a manual management mode.

Settings limitations

In order to find the optimal balance between price and quantity, the system needs space to experiment. For this reason, the less unnecessary limitations you set, the better your chosen strategy will work. For example, it's not advisable to set additional maximum CPC limitations when using the “Weekly budget” strategy.

Try to set appropriate parameters and not, for example, set too low of a CPC for expensive phrases or bid low when you have a large budget. This makes it more difficult to arrive at optimal indicators.

Restrictions are maintained over the course of a week, from Monday to Sunday. If you enable or relaunch a strategy in the middle of the calendar week, the limitations will be even distributed among the remaining days. If on Friday, for example, you launch a strategy with a weekly budget of 21,000 RUB, then 7,000 RUB will be spent each day on Friday, Saturday, and Sunday. Starting the next week, 3,000 RUB will be spent per day.

If your site (or app) users' activity changes quickly for reasons not directly related to to your selected strategy, it may take time to recalibrate, and the limitations you set might not be maintained. This can occur, for example, if you update the design of your site, add unpopular keywords to your campaign, or change the targeting.

Independent spending control

When you enable your strategy, think about whether it's really necessary to manage the positions and bids of individual ads in your campaign manually.It's better to redirect such ads to a separate campaign and use a manual strategy for them.

If you need to individually control your spending as part of an automatic strategy, it's better to set a weekly budget, rather than a maximum CPC. We also recommend that you base your average weekly budget settings on your ad campaign spending from the previous month. This way you can compare past indicators with current spending.

Keyword priorities

You can set a priority for each keyword on the campaign page. You can assign high priority keywords you think offer the best prospects. They will be moved into the best positions as much as possible and will be the last ones to be disabled when your budget runs low.

Assigning high priority to a keyword does not guarantee any impressions. This setting increases the importance of these keywords over others when allocating the budget.

Relaunching a strategy

After relaunching a strategy , the system always recalculates indicators, so that it can create a new forecast and select the best bids. This is done while saving and taking into account all statistics collected over the entire span of the campaign. The system takes the fact that these statistics are outdated into account: the newer data has greater “weight”. If the strategy was not in effect more than 28 days before relaunching, the statistics are reset.

Strategies are relaunched in the following cases:

  • changing the strategy settings
  • launching a campaign after having been stopped
  • topping up an account from a zero balance
  • converting the account to the payment currency.

Internal mechanisms allow the system to re-orient itself quickly, but it still requires some time to put together a new forecast. Daily expenses may exceed the limitations set on the day the strategy is relaunched.

automatic strategies
Rate this article
Thank you for your feedback!