Yandex direct is what you need to be an optimizer. Optimization of Yandex.Direct advertising campaign: how to add negative words, adjust bids, etc. So what - automatic optimization systems do not work

Hello everyone!

Campaigning Google adwords is not an easy process as well as. But, be that as it may, optimization contextual advertising (CR) is the most important action, without which the work of an Internet marketer does not make sense.

Surely, most of my readers had the experience of launching campaigns in both CR systems and, for sure, they did not get the desired results the first time. This happens to everyone, so don't be upset - it just takes time to complete the necessary work.

These same necessary work and there is the optimization, or maintenance, of campaigns, which we will talk about today. However, for a start, I recommend reading which ones we will use.

Have you read it? Then let's go!

Optimization stages

Optimization of campaigns in any contextual advertising system can be conventionally divided into several stages:

  1. Working with keywords;
  2. Working with other targeting parameters: geography, time, devices;
  3. Working with ads.

Thus, all the most important parameters of the RC are covered, on which the obtained result depends. Let's go through each, in order.

Working with keywords

At this stage, it is necessary to carry out several types of analysis and, on their basis, make any changes:

  • Analysis of search queries;
  • Efficiency analysis keywords;
  • Search and expand negative keywords.

In order to carry out this work, we need only one tab and three more nested in it. The tab is called "Keywords":

Let's start with the Search Queries report. Thanks to the data presented here, we will be able to understand what requests are entered by users of the search engine before clicking on our ad. In addition, this report expands semantic core (the list of keywords), if, of course, there are phrases that can be added to it, and the expansion of the list of negative keywords, if there are such queries that need to be cut off.

Working with the data in the report is easy:


Thus, new phrases for contextual advertising are searched for in the "Search queries" report. The expansion of the list of negative keywords also occurs in almost the same way - only you will need to click on the "Add as negative keyword" button and leave only those words of the phrase that are its basis.

As an example, the phrase “how many minutes to sterilize jars in the microwave” is not at all target, so impressions on it should be cut off, you never know. However, the words “how much” and “microwave” can potentially bring customers, therefore we cut off the word “sterilize”. To do this, it is enough to leave the above word in the "Minus word" field (sorry for the pun):

Also, remember to select the level where the non-targeted query is added. You can add the rest of the query words in the same way, or through the "Negative words" tab.

So with just one report, we were able to find new phrases and cut off non-targeted visits. Therefore, we will spend a little less on advertising, and if the costs increase, they will become more efficient.

Now you should analyze the keywords already in use. To do this, we will use the "Keywords" tab, where all the qualitative indicators of the elements of the semantic core are presented:

For example, if a key phrase has poor performance (low conversion rate, low CTR (interaction rate)), then you should disable it, or correct it.

By the way, in order to track the conversion of advertising campaign elements, you need to set up this tracking. I already wrote about this, read it.

Working with other targeting options

Here our task is to identify the most effective display regions (if you are shown in several regions) and types of devices (suddenly you have mobile traffic converts much better than desktop). For this task we need two reports located in the "Settings" tab:

Let's start with geography. Here you also just need to analyze the quality indicators in certain regions or n / a of our country:

If a region / city shows good results, then it makes sense to apply an upward adjustment for it. So, for example, in one client account the city of Ufa turned out to be the most efficient, for which an adjustment for an increase was applied, which gave the result in the form of a slightly larger number of applications from this city. Unfortunately, I cannot attach a screenshot, since now there is no such possibility.

With devices, everything is a little simpler: just go to the "Devices" tab, analyze quality indicators, apply the adjustment:

This is the simplest way to optimize campaigns in Google Adwords based on working with targeting parameters.

Working with ads

At this stage, we will work with elements of advertisements:

  • Title;
  • Text;
  • Display URL
  • Extensions.

Before you start working with ads, analyze their quality indicators and write them down somewhere, so that after all the changes are made, measure their improvement or, on the contrary, deterioration.

Title

The ad title should be as relevant as possible to the user's search queries. To do this, you need to specify in it the main keyword of the ad group for which impressions occur (if your group has several keys):

Now there are no problems with specifying the necessary information, since Google has made certain additions in the form of a second title and an elongated description (text) of the ad. So you can make your headline really big, which will help attract the attention of more potential customers.

Text

The text should clarify your proposal, which we announced in the title. Please indicate additional information about a sentence using words from a key phrase. So these words will be highlighted in bold when they match the user's request, and the more bold he sees in the text, the higher the likelihood that he will click on your ad and buy from you.

Optimization of an advertising campaign is almost the most important point during its implementation, on which the financial result of the promotion depends. Many advertisers believe that it is enough to work out a high-quality context at the start so that it works well and makes a profit. However, it is impossible to reliably predict the course of the campaign, so regular optimization - an integral part of successful promotion.

The main task of the analysis and optimization of advertising campaigns (AC) is to increase the financial return on advertising. This can be achieved by acting in two directions - increasing the number of targeted actions (conversions) on the site (landing page), and at the same time, reducing the cost of the targeted action.

Two main tools are used for the analysis of advertising campaigns - Yandex.Metrica and Statistics in Yandex.Direct... Data - this is due to different approaches to calculating site traffic and accounting for targeted visits. Metrica counts visits (visits lasting more than 15 seconds), and Direct counts clicks.

A conversion to Yandex.Direct is a transition within which at least one conversion was made. The metric counts the number of completed conversions (if the landing page provides for several targeted actions).

Considerstructure of your campaign. If the campaign is built on the principle of “one key, one ad”, it is enough to analyze either the ads or the keys. If according to the principle "one ad - a group of keys", then the estimated indicators should be viewed both by ads and by individual phrases, since they can differ significantly within one group.

Optimization of Yandex Direct on search: key indicators and working with them

    Number of impressions;

    Bounce rate;

    CTR;

    Number of conversions;

    Target price.

    Refusals

A refusal is a visit that lasts less than 15 seconds. If the visit ended with a refusal, this indicates that the ad interested the user, but on the site he saw something completely different from what he expected. Failure analysis is most conveniently done in Metrica (Direct Summary Report), since you can immediately establish the root cause of the high bounce rate - garbage traffic.

It is optimal to review all the phrases for which refusals occurred daily. Evaluate search phrases and immediately add them to campaign or ad group negative keywords. If a campaign with broad semantics, or you are analyzing a campaign for certain period, it is convenient to use the filter in order to cut off phrases with a small number of clicks. If only 1 transition was made for a phrase with refusal, then the% of refusals for this phrase will be equal to 100%, which, of course, is not a reason to disable the phrase and any conclusions. There can be quite a lot of such phrases, so it is worth choosing phrases with a sufficient number of transitions for analysis. I start looking at phrases with 10-15 clicks.


Overall campaign bounce rateshould not exceed 20%.

The second reason for refusals is irrelevant ad and landing page. The user saw not at all what he expected.

When the keyword is relevant to your topic, ad, and landing page, but consistently shows a high bounce rate, the keyword is better. exclude from the campaign. But before excluding, be sure to make sure that it doesn't convert. This situation is possible when the phrase is wide, and the ad is written not for a narrow segment of the target audience, but for several segments at once. Therefore, some of the target audience leaves the site, and some of them make conversions. At the same time, the cost of conversion should be acceptable.

In any case, it is very important to analyze failures usingwebvisor.

An example of a campaign with high-frequency phrases that have a high bounce rate, while converting well.


Scheme of work with a bounce rate.


  • Impressions

One of the problems affecting the number of clicks from ads is the insufficient number of impressions for the key phrases of the campaign. If you are using medium and high frequency keywords in your campaign, but the number of impressions is very small, you are likely to be tooThe campaigns were “minus”, or the impressions were in guarantee and dynamics.

In the first case, it is recommended to look through the list of negative keywords again - sometimes keywords accidentally fall into it. In order to roughly estimate the impact of negative keywords on the number of impressions even before the launch of advertising, you can use the Budget forecast service, where you can specify the display region, a list of key phrases. Then compare the number of impressions for keywords with and without a given set of negative keywords.

In the case of a significant number of impressions in the guarantee and dynamics It is recommended to increase your bids, as well as the CTR and quality of your ads, as these are the parameters that affect your placement. Perhaps, there were a lot of impressions, but if the user did not view the page with results until the end (i.e., did not view the guaranteed impressions block), it is considered that there was no impression.

During the campaign, you will find the keys that bring the bulk of the conversions. Therefore, if you are on a limited budget, you can only raise rates for those keyswhich are extremely important to you.

An example of estimating the number of impressions by line item in Yandex.Direct statistics.


Scheme of work with the number of impressions whenoptimizations Yandex direct


Extremely important indicator campaign quality - CTR - click-through rate. CTR directly speaks about the quality of ads, the correct selection of key phrases in accordance with the interests of the audience, as well as the attractiveness of the offer depending on the needs of the target audience and competition in the niche. It's helpful to create more than one ad to identify the most clickable option. But you shouldn't get carried away with the number of ads in a group in search - traffic and clicks may not be enough for statistically significant testing.

At the initial stage of the campaign, a CTR value of 10% or more indicates that the campaign is initially well-developed and has good prospects. As the campaign unfolds and is further optimized, you can achieve a CTR of 15-20%. To do this, the first step is to analyze the click-through rate of the campaign in the context of ads and identify those that do not have a high enough CTR.

When analyzing, you should pay attention to the ads, the number of impressions of whichexceeded 30, while there was not a single click. It is too early to draw conclusions with fewer impressions.

Be sure to consider the display position - in guaranteed impressionsthe click-through rate is lower than in the special placement - on average 10 times (as can be seen from the example from the previous paragraph). In addition, search sites - Yandex partners can distort the statistics, therefore it is better to fix the position "Special placement", the name of the site "Yandex" in the filters.

An example of setting up filters to identify ads with low CTR.


When analyzing nonclickable ads, it will also be usefulthe "Search queries" statistics report, which shows the user's direct query and the key phrase of the campaign for which it was shown, but there was no click on the ad. Using the filter, it is necessary to fix the number of the group or ad, and analyze the relevance of the search query and key phrase.

At this stage, you can understand the reason for the low attractiveness of the ad for the user, identify additional negative keywords, and also refine the key phrase (add a lower-frequency key to the campaign).
The main criterion for the analysis of conversions is target price. Before starting a campaign, it is extremely important to calculate the maximum cost of a targeted action. The target price is the maximum cost of the target action performed on the site, at which the advertising will be profitable. You determine the value of the desired profit yourself, based on the parameters of the business and the level of competition in the niche.

Optimization is about increasing efficiency. As applied to contextual advertising, this is a decrease in the cost of a conversion (call, order on the site) or the achievement of more complex goals.

Optimization is possible if two conditions are met: there is a way to measure the cost of conversion and the ability to change the parameters of advertising campaigns.

Optimization is not a one-time action, it is an ongoing process:
Analysis of statistics\u003e Formulation of hypotheses\u003e Design of the experiment\u003e Analysis of results.

What is conversion quality

The quality of clicks from an ad in contextual advertising is determined by the conversion from these clicks. In contextual advertising, you can control the choice of phrases (show / not show), ad position and display time. Ads texts are given to computers only for sites with a large (tens of thousands) and constantly updated range of products.

The campaign advertises different products or services. Some goods always go better, some worse. It turns out different conversions for different products. Several phrases are responsible for advertising one product - group... An easy way to optimize is to identify conversions by group and shift your budget from low converting to high converting groups.

The phrases in the group are different from each other language... The phrase may be closer to the purchase and further. The phrase “buy a bike” is more conversion than “bicycles”. Conversion phrases are often low-frequency. Let's remember this for the future.

Usually the higher position of the phrase - the more high-quality visitors come through it. This has a simple explanation - the more the customer is determined to buy, and not to “pick”, the more likely he will click on the first or second ad and will not search further. Of course, there are always exceptions, but they are insignificant in terms of volume.

From show time the conversion of visitors depends quite strongly. This dependence has a unique profile in each market. But usually the highest conversion in working time from 10 to 14, and then until seven o'clock it goes down. 100 clicks from 10 to 11 will generate more sales than 100 clicks from 19 to 20.

Automatic optimization problems

The idea of \u200b\u200bentrusting the optimization of campaigns to a robot is visited by any programmer faced with contextual advertising. Optimization problems are the most efficient area of \u200b\u200bapplication for computers. A lot of contextual advertising optimization services have grown from this idea. Each major advertising agency there is such a service. But the way programmers and clients saw in their dreams the work of such services (by pressing one button), they do not work. Why?

Let's imagine that we set up ad impressions for two phrases and the following statistics came up in a month

Can we draw a conclusion on the basis of such data about which phrase is better? Obviously not. Too few impressions, clicks, and conversions. And how much is needed for an informed conclusion? If you hold the statistics in a corner and ask this question for a long time, the answer will be about 35 conversions. With such a volume of impressions per month and conversion, let's say 5% (which is a lot), it will take 165 months to collect statistics. And often, as we noted earlier, conversion phrases are low frequency. Dead end.

Of course, there are more frequent phrases in the campaign, and the stability of indicators in Yandex.Direct makes it possible to ignore the strict laws of statistics and turn off the display of a phrase in the absence of conversions with just 300 clicks. Programmers come up with clever algorithms that try to predict future conversions by converting these phrases on other sites, by user behavior, and by the phases of the moon. But you can't argue with mathematics - at least somehow automatic optimization works on high-frequency phrases and requires a long campaign period to collect statistics.

Another problem with automatic optimization is that it does not know how to select CONVERSION additional phrases. For example, to the word “buy two-suspension” there are additional words “buy a frame” and “shock absorber for a bicycle”. Their conversion rate depends on the availability of these products in the store, which the automatic system cannot find out about. And it will take years for her to draw a conclusion based on statistics.

So what - automatic optimization systems don't work?

Nevertheless, automatic optimization can and does bring serious benefits - but only for large (tens of thousands of phrases) campaigns with a budget of 500,000 rubles per month, where many high-frequency phrases are used and only works in tandem with a talented analyst who understands where to send system, and where it is not worth using.

The closest approach to the approach of symbiosis of automatic systems and a qualified analyst came in Calltouch. There the system proposes solutions, and the analyst accepts them or not. The system is now undergoing closed testing. The worst came up in Yandex, where decision-making is not transparent, and for the assessment it is suggested to wait “6 weeks or more”.

For small campaigns, the most effective method - setting up accounting for the cost of conversion using the metric and the same Calltouch, and then manual analysis and decision making by the analyst.

We are now following this path in Eureka, automating only the receipt of information for decision-making.

Types of display strategies in Yandex.Direct

The simplest one is highest available position... You yourself determine the maximum bid for a specific phrase. Yandex displays the phrase at the most accessible position for this bid. In essence, this is not a strategy, but a direct way of working with bidding for ad display locations. We use this strategy in conjunction with the bid management robot for all campaigns.

Display in a block at the lowest price - Theoretically, it allows you to hang in a special placement for minimal money. In practice, savings do not come out due to the effect of warming up the auction, when two advertisers with such a strategy meet on the same phrase. Each ad strives to enter the block and progressively higher bids are awarded.

Average cost per conversion - an attempt to use automatic optimization of an advertising campaign by Yandex. It has all the limitations described above, combined with absolute opacity in decision making. Imagine that you are driving a taxi, the driver is not familiar with the city and wanders, and you are blindfolded and cannot give directions, although you are driving through familiar places.

Average return on investment (ROI) - the same as the average conversion price, but for online stores. The approach itself can give interesting results with painstaking and correct setting campaigns. But again, the lack of transparency in decision-making does not make it possible to effectively use this tool. More benefit will come from manual analysis.

Average cost per click - Yandex is trying to catch up with more clicks by keeping the average price close to the one you specified. The situation is the same as in the average conversion price, but now the taxi driver does not even know the final destination of the route. Yandex will choose phrases for you that bring more clicks with no matter what kind of conversion. It will also collect cheap clicks - from low positions and at low competitive times - just when the conversion for them is minimal.

Independent management for different types sites - this is not a strategy, but a switch that allows you to separately set strategies for displaying ads on search and YAN.

The rest of the strategies are combinations of the above.

All these strategies have unpredictable results. The only justification for their existence is that this result is usually slightly better than randomly placed bids and selected phrases - the typical state of a campaign assembled by a person with no serious experience with PPC advertising.

How bidding for places in Yandex.Direct work

In Yandex, bidding for ad space is based on the rules of the Vickrey Auction (VCG). The player who named the highest bid wins the auction, but pays not it, but the price named by the second participant plus the bidding step. Such an auction is better than the usual one in that it is profitable for the players to name their maximum bids, and not to consistently raise them “playing on their nerves”. Thus, the auction can be held in one move.

Players are expected to accurately calculate their maximum bets. IN real life this does not happen, but if the auction is held once, then everything works in accordance with the theory. Everything changes if auctions are held regularly - players have a chance to adjust their bids with an eye to the actions of other players. The theory of the Vickrey auction is no longer designed for this, and the strategy of announcing the maximum bids ceases to be dominant (always bringing the best result).

When bidding for positions in Yandex.Direct, practice is even further from theory - trading takes place continuously, players' bids change independently in time from each other. Most of the bets are made based on inaccurate data on the rates of other players. Players have different goals, different qualifications, different methods of calculating bets, a lot of mistakes. There is no theory describing the optimal behavior of players in such a situation. This means that the existence of an “optimal automatic strategy” is also impossible.

How a Bidding Robot Works

For campaigns controlled by a robot, Direct is set to the “highest available position” strategy. In the robot's interface, you can set the maximum bid, the main strategy and the reserve one for each campaign and even for each phrase, which is triggered if the maximum bid is not enough for the main one.

The point of the robot's work is to hide your maximum bet from other players... For example, if you set the strategy SP + 0.01 in the robot, it will place a bid on a phrase every five minutes at the entry price of the special placement + 0.01 cent. If the third player in the special placement uses the strategy “display in the block at the minimum price”, then by consistently increasing bids per working day, the robot will raise the bid by a maximum of $ 1. (in 8 hours he will make 96 bets). If we had simply chosen the strategy of “display in the block at the lowest price” in Yandex.Direct, the price would have skyrocketed to the maximum bid of one of the players, at which the second player would have received all the clicks for that day.

Benefits of using a robot

  • You do not tell other players your maximum bet. Sincerity is not the dominant strategy in successive Vickrey auctions, so by not bidding you get the best results possible.
  • Rates increase at a controlled rate.
  • You can choose very different main and backup strategies
  • You can define your own strategies for individual words without separating them into a separate campaign

We have released a new book “Content Marketing in social networks: How to get into the head of subscribers and fall in love with your brand. "

Many advertisers are sure that advertising specialists after launching advertising campaigns do not do anything there - they just sit, count the number of clicks and spend their money, spitting on everything else and never returning to these campaigns. Oh, if only.Optimization of advertising campaigns Yandex.Direct should be done regularly, only then the advertising will be effective and generate income.

Today I want to tell you what happens to an advertising campaign after you click the magic "Start" button.

Backing track

It would seem that the campaign has been launched, there are impressions, clicks too, but the conversion is a little lame, or in general there are a lot of impressions and few clicks. The first thing that comes to mind is to check the requests for which ads are displayed. To do this, go to the "Direct - Summary" report in Metrica:

and select the appropriate groupings that interest us:

In this case, this is a slice by campaign, ad, keyword from Yandex.Direct and the search phrase for which the ad was shown:

For example, in our campaign we have a query "makeup school". Having built the required report, we see that our ad was shown not only for the query "makeup school", but also "why is it impossible to make up for school." In this case, the words "why", "not" are necessaryadd as to the phrase or the whole campaign.

Why is this needed? Wordstat statistics are constantly updated, users are constantly generating new queries that are not in the system and that you could not even think of. Therefore, you do not need to think that, having eliminated everything once and for all, you will no longer have to return to this.

Assessing your ad positions

If it seems to you that you are not getting enough traffic, you need to look at exactly where your ads are shown - in a special placement or in a guarantee. To do this, you can go to the standard reports in Yandex.Direct and check the boxes that you need, not forgetting to click opposite the "Position" slice. As a filter - "Site type" - "Search".

Why is this needed? You can estimate how often you lose by. Perhaps only a slight increase is needed to minimize guaranteed impressions. After all, whatever one may say, the special placement attracts more attention of users and, as a result, gets more clicks. But it all depends on your budget and capabilities.

Evaluating the effectiveness of phrases

It's not a secret for anyone that often some phrases bring long-awaited orders, and some just drain the budget. If there are no non-targeted impressions for them, in general everything is fine, but there is no effect, then you can:

  • try to rewrite the ad and evaluate the effect;
  • lower the rate for these requests;
  • reluctantly, refuse these requests.

If the phrase is effective enough and brings a conversion, you need to find out if you can get more clicks on it. To do this, see if it often falls into the warranty and if you can somehow fix it. Perhaps all that is missing is a small increase in the rate, which will then pay off with interest.

Exclusion of sites in YAN

Why is this needed? Often, ads are displayed on sites that are completely unrelated to your ad. As a result, idle impressions lower the notorious CTR, and ineffective clicks waste your budget.

To understand which sites you don't need, use the following reports:

  1. Report in Yandex.Direct. Here you can estimate costs by type of site and its effectiveness in terms of impressions, clicks and costs.

How is it useful? You can upload the report, sort the sites according to the criterion you need, select those that should be excluded and add to "" in the campaign settings.

  1. Report in Yandex.Metrica. In the menu "Reports" - "Standard" there is a report "Direct - sites", which can also be segmented as you need. This is how this report looks standard:

How is it useful? In Metrica, you can segment the report according to the goals you need, and, therefore, evaluate the effectiveness of the site not from the cost side, but from the side of its effectiveness.

Additional adjustments

Among other things, you can quickly conduct a quick analysis in Yandex.Metrica and configurebid adjustments in Yandex.Direct.

  1. Adjustment for gender and age. You can use the Age and Gender reports to customize this adjustment:

Here you can estimate who comes to your site by age and gender, estimate who makes the most conversions, and adjust the adjustment of bids by gender and age. For example, I see that my target audience is women from 18 to 34 years old. Then I go to the settings of my ad campaign and raise the bid based on this data:

  1. Correction on mobile. To configure it, you need to go to the "Devices" report:
  • Adjustment for visitors to the site. To set it up, you must have conditions set up. To do this, you can select from the existing retargeting conditions or click on the link "Setting conditions":

    • And add a new condition:

    As seen, optimization of Yandex.Direct campaigns requires an assessment of what you are doing and a clear understanding of why you are doing it. Therefore, if in doubt, you should contact a specialist, and everything will definitely be fine.

    Below is a case report on optimizing advertising campaigns for one of my clients who sells a franchise.

    However, I am posting this case here not in order to show its “mega efficiency”, but in order to demonstrate the LOGIC, which should be followed when optimizing advertising campaigns.

    Of course, this case is not exhaustive - there are many "little things" and "tricks" that cannot be described within one report.

    But, nevertheless, this report forms a general idea of \u200b\u200boptimization, and I am sure that after a detailed acquaintance with it, you will know exactly the answer to the question: "WITH what I should start optimizing Yandex.Direct advertising campaigns. "

    So, the case.

    Studia12 ad campaign optimization report

    Before starting to optimize directly, I analyzed the statistics for the past year for all your advertising campaigns.

    Specifically compared period A (2nd half of the year, ) and period B (1st half of the year, from March 27, 2016 to August 28, 2016)

    It turned out that the average conversion price in the second half of the year (Period A) was 405 rubles. and turned out above almost 2 times than in the first half of the year (Period B). (In the first half of the year, the average conversion price was 248 rubles.)

    Those. it is obvious that the “drawdown” for all indicators began precisely in the second half of the year.

    Therefore, it was decided to optimize advertising campaigns based directly on statistics for the second half of the year ( from 28 August 2016 to 27 March 2017).

    Optimization step by step

    I must say right away that they are listed below are common steps towards optimizing advertising campaigns. Those. describes those actions that "at a distance" will have a positive effect on the overall conversion.

    I preferred to omit various subtleties and nuances in order not to overload this report with them.

    1. First of all, I looked at the general statistics for all advertising campaigns in order to determine which campaigns should be continued to work with, and which should be immediately removed from the promotion, so as not to overload the interface and complicate further analytics.

    As you can see, there are several campaigns in the promotion, the target price for which ranges from 800 to 10,000 (!) Rubles.

    Why do we need SUCH campaigns? If others are much cheaper ?!

    In general, remember a simple rule: if an ad campaign (especially YAN) after 100,000 impressions shows negative results (high cost per lead, low CTR and conversion), then it is better to stop such a campaign , in order to avoid various misunderstandings.

    Therefore, the first thing we do is turn off all those advertising campaigns that have gained 100,000 (or more) impressions and showed negative results. From now on, we DO NOT LAUNCH these campaigns.

    Why don't we launch it?

    Because these advertising campaigns simply attract NOT OUR target audience... And no matter how we “bother” with A / B tests of texts, pictures, disconnecting sites ... if the audience comes not ours, then the money "merges" in vain.

    Here are the names and IDs of the campaigns that have been paused:

    p_business_RSI_by_all 17205885

    p_business_RSI_az_all 21998537

    p_franchise_RSI_by_all 17206493

    p_franchise_RSI_az_all 21998547

    p_franchise_RSI_am_all 17206498

    p_product_RSI_ru 12028870

    I strongly discourage launching and reloading these campaigns in the future.

    2.1 Now let's look at the statistics for specific ads. We are primarily interested in those ads (or rather key phrases), the cost of the goal for which was more than 500 rubles:

    For example, on request: small business ideas the cost of the target was 4416 rubles, which, you see, is "not cheap".

    Thus, we should go through the entire list of keywords and just turn off everything that "eats" the budget. At the same time, I note that after we turn off the "bad" ads, the "good" ones will start typing more impressions. However, at first I do not recommend increasing the cost per click on “good” ads, at least not on YAN.

    Below are some identifiers of the worst ads (I turned them off):

    M-1793463673

    M-1826407415

    M-1826407424

    M-1826407031

    M-1826512930

    M-1826408346

    M-1826407424

    M-1826513575

    M-1826407389

    M-1826408421

    M-1826512849

    M-1793437979

    By the way, queries like: looking for a business gave on YAN about 300-400 clicks with zero conversion.

    Thus, a general list of more than 500 phrases and ads was worked out.

    2.2. “Automatically added phrases” also ate a certain part of the budget (they gave 2 conversions per 10.000 rubles of the budget):

    But we will deal with them when we analyze the missing negative keywords in the search.

    3. For now, let's move on to the sites. We will filter the sites according to the same principle by which we filtered key phrases. All those sites, the cost of the goal for which was more than 500 rubles, we will add to the "black list".

    Why did I take exactly 500 rubles for the lower bar?

    Because if you take 300 or 400 rubles for the lower limit, then you can very much cut coverage audience, which, in general, is highly undesirable.

    In addition, instead of our “bad” sites, Yandex can “slip” new ones, the quality of the audience on which will be, frankly, “not top”.

    Therefore, those who are too pedantic about the selection of sites on the YAN often fall into trouble - instead of increasing conversion, they often notice its deterioration. They do not understand the simple fact that instead of one “bad” site, Yandex can slip 10 new, shitty ones (sorry for the pun!).

    We are not we will repeat the mistakes of others and approach the selection extremely intelligently.

    Here are some of those sites that will definitely have to be disabled:

    Note that a total of 82 sites were identified that negatively affect conversion. These sites will be included in the list of "prohibited" for each YAN advertising campaign.

    Complete list of "bad" sites

    4. A similar analysis was carried out by region. After analyzing the reports, the most expensive regions were identified and disabled:

    Why exactly 600 rubles. is it taken here as a starting point?

    Because when filtering regions, you should be a little more liberalso that do not lose coverageand the “quality” of the audience does not depend as much on the region as on the same advertising platform.

    Note that the regions will have to be adjusted in the parameters of EACH YAN advertising campaign. Yes, it's a little inconvenient and energy-consuming, but effectively.

    5. Now with regard to the negative keywords. In your case, Yandex has added a bunch of garbage to search campaigns through "automatically added phrases" and synonyms, withCTR near zero:

    If you advertise by phrases in broad correspondence, then you constantly need to track what REAL queries are showing (you can see this in Report Wizard -\u003e Search Terms). Otherwise, a huge part of the budget will be wasted! (No exaggeration.)

    Of course, I went through the list of requests and removed unnecessary phrases, but from now on, requests need to be tracked on regular basis.

    This is another reason why the budget is shooting milk.

    Creation of advertising campaigns for YAN with image ads

    As we agreed, for YAN, I created new advertising campaigns containing image adsas well as the most effective key phrases.

    Russia

    Armenia

    Kazakhstan

    Kyrgyzstan

    You have separate campaigns as well as unique landing pages.

    Consequently, a unique YAN advertising campaign with image ads was created for each region.

    Now I'll tell you a little about the creation of these campaigns themselves.

    First of all, I chose those key phrases that provided the greatest influx of conversions in YAN, at an adequate average cost.

    Some of these phrases are:

    franchise business

    business for beginners

    small business ideas

    ideas + for small business

    open your business ideas

    "start business"

    new business ideas

    best business ideas

    Here is an example of an ad (format 240x400):

    After that, for each of the previously listed regions, a uniqueadvertising campaign with image ads.

    This is enough to test the effectiveness of a graphic format in your niche.

    Summary

    However, this does not mean at all that tomorrow the consumption of your advertising budget will fall by 2 times ...

    The real result in the form of an increase in overall conversion (decrease in the cost of the goal) will be noticeable at a distance of about a month.

    This is especially true of search campaigns, which, as it turned out, mostly broadcast ads at all not that the audience you needed.

    On this, perhaps, everything.

    I hope this report was useful to you.

    Sincerely, Dmitri.