Advertising in Yandex.Direct or Google AdWords should solve specific business problems. To determine if you are moving in the right direction, you need to monitor each RK, optimizing its settings and gradually increasing the flow of targeted traffic. Customizing your ads target audience and budget, after a certain time after its launch, evaluate the results and increase ROI by comparative analysis multiple campaigns.
3 questions before starting work
What are the goals of your advertising campaign?
All goals should be expressed in measurable terms:
- number of target actions;
- CPA — cost of action in rubles;
- campaign conversion percentage.
Let's say if you invite to a webinar, then your optimization goal might be to get 200 registrations. Sell flower arrangements - increase the number of orders from the site by 4-5 times. Success in achieving results is determined by KPI. This is exactly the indicator that determines the whole. You need to make comparisons on it even before making all the settings.
What tools to use for analytics?
Best suited for this type of problem.
- Google Analytics.
- Yandex.Metrica.
- liveinternet.
Why consider the effectiveness of contextual advertising?
Without evaluating conversion rates, you cannot make adjustments to the course of an advertising campaign. Ultimately, the results can be extremely unexpected, quite possibly unpleasant. Without conversion assessment, it is impossible to determine ROI - an indicator of your investment effectiveness. Moreover, it is important to calculate the conversion for individual advertising campaigns. This is the only way to understand what brings the desired result, make the maximum list of negative keywords, set up the necessary extensions, and then launch effective retargeting.
Analysis and conversion calculation
First of all, we fix the initial conversion rates. Then we determine the campaign budget and the average cost per click. Start from Yandex.Metrica data obtained during the previous optimization period. You can take information for all time and for the last week. So your audit will be deeper, and your expectations will be more justified.
Once you have decided on the time frame, move on to cleaning the campaign from non-targeted clicks. What should be removed first?
- Inefficient keywords (everything that brought no more than 3 clicks).
- Garbage words (what was skipped from Wordstat).
- Ineffective display regions (characterized by low conversion).
- Inefficient platforms in YAN.
For cleaning you will need statistics:
- by keywords for the period of interest;
- according to the phrases of YAN;
- GEO for campaigns covering several regions or the whole of Russia.
Reports are summarized in an Excel spreadsheet. Now your task is to eliminate everything inefficient. After cleaning, we proceed to the analysis. It is carried out 1-2 weeks after the optimization. The number of clicks, CTR, cost per click and conversion rates are compared - before and after optimization.
Upon completion of determining the conversion values, we display the KPI. We use for this the conversion indicators obtained after cleaning. For example, your conversion rate is 5%. This means that out of 20 attracted visitors, only 1 will leave an application. Let's say average check- 2000 rubles, and the profit from the sale with this number of visitors will be equal to 400 rubles. Accordingly, you can spend a maximum of 400 rubles to attract 1 client (or 20 visitors at the current conversion level).
Let's assume that the marginal cost of a click will be equal to 20 rubles (400/20). If a click costs $20, attracting 20 visitors will cost $400 (20X20), and the revenue at the end of the advertising campaign will be $1,600 (2000-400).
Conclusion: the maximum cost per click depends on the amount of net profit per client, conversion and how many times you want to increase investments. A click price of 20 rubles will allow you to return the money spent on advertising, but nothing more. A 2-fold increase in return on investment is possible if marginal price click will not exceed the amount of 10 rubles.
Measuring ROI
After measuring KPI, we will be able to determine the most important parameter of any advertising campaign - ROI. This is how we know the true value of our investments. The following formula is used to determine ROI:
The calculation will look like this:
((2000-1600)-400)/400 = 0.
The advertising campaign was not successful, because the ROI is zero. The investment paid off, but it was not possible to make a profit. To achieve a positive result, the cost of a click should not exceed 10 rubles. Only then the ROI will be 100%, and you will earn 2 times more than you invest:
((2000-1600)- 200)/200 = 1, or 100%.
Rules to follow when calculating the effectiveness of contextual advertising:
- Your investment should pay off. And advertising is to make a profit, and not just cover investments.
- In your measurements, you must be accurate and consistent. Cleaning and performance analysis first, not defining KPIs.
- The figures must be converted into actual sales. Otherwise, there is no point in starting calculations.
- Analyze both your successes and failures. Take into account the worst and best campaigns, compare them by all indicators.
- Any analysis must be honest and objective. Don't sugarcoat the results. Only in this way can you achieve a truly excellent result.
The headache of any person who decides to use Direct is evaluating the effectiveness of advertising. There are impressions, there are clicks, there are even leads. But is it too much or too little? Are all campaigns working effectively? Isn't the budget going to waste? To answer these questions, you need to analyze the work of the context, and on its basis already think about how you can squeeze more out of it. I dedicate today's blog entry to this difficult, but really touching topic.
Evaluation of the effectiveness of contextual advertising
To evaluate something, you must first determine the goal, the main performance indicators contextual advertising, comparison objects/periods and data acquisition tools.
Goals
There should always be specific and measurable goals. Not abstractly “increase sales”, but, for example, “provide 100 leads”, “achieve an order value of no more than 200 rubles”, “conversion from 15%”, etc.
Without this, there will be no point in watching something yourself, and a misunderstanding will arise with the contractor, and there will be empty disputes about who had what in mind.
Now about the indicators of the effectiveness of contextual advertising.
Indicators
Conversion
The target action when analyzing the effectiveness of contextual advertising can be:
For many, this is the main data used, although it is not fully expressive.
Economic
The most significant indicators of the effectiveness of contextual advertising - directly reflecting the contribution of Yandex.Direct to the income from the site:
- ROI - return on investment. ROI = (investment income - investment size) / investment size * 100%;
- CPA is cost per action. CPA = amount of costs per context / number of targeted actions;
- CPL is cost per lead. CPL = amount of advertising costs / number of leads;
- CPO is the cost of the order. CPO = amount of expenses / number of confirmed orders.
These calculations will allow you to see the picture more clearly, in terms that are understandable to the business.
Behavioral
The characteristics of visits can also be a criterion for evaluating the effectiveness of advertising.
For infotainment projects, they can be practically the main criteria, for commercial ones they are also significant, because behavior correlates with conversion.
What to compare
The question is not so simple, with what to compare the accumulated data:
- with last month;
- with the corresponding last year;
- for the same period with other channels;
- with the advertiser's expectations, etc.
The first option, which allows you to evaluate the effectiveness of contextual advertising, seems to be the most obvious, but it does not take into account that in some time period there could be something that could greatly affect the result: holidays, an advertiser's promotion, a collapse in prices from competitors, or just some event that distracts the minds (elections, the Olympics, etc.).
The second one is even more difficult, because there could be strong changes in campaigns, on the website, in assortment, prices, etc.
The option of evaluating the effectiveness of contextual advertising in comparison with another source makes it difficult to choose the optimal channel, and you also need to know all the actions taken on it.
The latter threatens to be vague and insufficiently based, because you need to know what the plan was made of, how real it was and how it met the needs.
Tools
You need to select the source for receiving and processing data. Of course, Excel and various uploads are most often used.
How to increase the effectiveness of contextual advertising
When goals, KPIs are defined and data is accumulated, it is necessary to start analyzing and adjusting campaigns.
To increase the effectiveness of contextual advertising, it is advisable to move from large entities to a detailed examination of their components. That is, first we evaluate the effectiveness of the context as a whole, then a separate campaign, group, ad, phrase.
- We compare the dynamics of KPI achievements in different sections across the entire channel, between different campaigns and within each.
For example, here you can see that Direct works much better than AdWords, where there are obviously serious problems or errors in utm tags.
- We identify critical points and try to guess what could be the cause. We check if the problems lie outside of contextual advertising: an inconvenient / ugly site, poor-quality processing of applications, the wrong phone number on the site, incorrect substitution in call tracking, etc. We look at what changes were in advertising: money ran out, a new option was tested text, there was a rate increase, etc.
- Let's examine the data point by point. We look at which phrases brought leads and conversions, how much they paid for them, and whether this price suits you. We check the expenses for phrases that did not generate income.
Example: look at leads for specific ads. 
- We redistribute the budget for key phrases that provide more profit.
- If, according to the effectiveness of advertising, it is clear that the costs are too high, we are looking for ways to reduce them.
- We check the adequacy of the phrases for which there were impressions. If necessary, we correct: we transfer non-target keywords to negative keywords, good options, if possible, can be added as keywords.
Example: here are impressions for non-targeted phrases. 
The landing page is ready, contextual advertising is running, but there are no orders and the phone does not break from calls. You have analyzed your site, checked its performance on different devices, and made sure that the problem is not in the USP. So what's the matter, why advertising is not profitable?
Where's the money?
This is because the context is not a golden antelope, it does not produce money by itself. This is a tool, and as you know, all tools require tuning and skillful handling. Our short guide to contextual advertising analysis will help you answer the following questions:
- What metrics are needed to work with contextual advertising and how to analyze them;
- How to reduce the cost of customer acquisition and improve contextual advertising performance;
- How to achieve the most efficient contextual advertising.
Collecting data...
Before analyzing anything, you need to take care of collecting data. The main analytics systems are Yandex.Metrica and Google Analytics. Thanks to them, we not only see all the traffic data, but we can also track user behavior:
- they stay on the site or immediately leave (bounce rate);
- how much time they spend on the site (time on the site, depth);
- what actions are performed or not performed (goals).
In turn, gives us all necessary information about calls, linking the call from the visitor to his session and advertising source.
Both Yandex.Metrica and Google Analytics have their advantages and disadvantages. Their functionality is basically the same, however, the data often diverges. Therefore, for complete analytics, it is necessary to use both of these systems, which is not always convenient.
With the help of the end-to-end analytics service, the task of analyzing the effectiveness of contextual advertising is greatly simplified. For example, the CoMagic analytical platform independently collects, combines and transforms data from different sources, displaying them in a single interface.
All analytics is presented in a convenient end-to-end reporting format, and all you have to do is optimize the advertising itself. We'll talk about this a little later, but for now: the counters are set, the goals are set, call tracking is enabled - we begin to track performance indicators.
Main characteristics
There are quite a few performance indicators: CTR, ROI, CPO, CPL, LTV, CAC, DRR, etc. It is easy to get confused in them, so let's simplify the picture and look at the key ones.
1. CTR (Click through Rate) - the percentage of your ads being clicked through. The calculation is simple: we take the number of clicks and divide by the number of ad impressions. The higher the click-through rate, the better tuned advertising campaign: The cost per click will be lower, and ads will be shown higher and more often.
2. Conversion - the ratio of visits to targeted actions: call, form callback, using chat, etc. to the total number of visitors to your site.
3. CPA (CPL) (cost per action/cost per lead) – we divide the advertising costs by the number of target actions, we get the cost of one action (or a lead – an attracted user). The lower this indicator, the more profitable advertising is for you.
4. ROI (return on investment) - the return on marketing investment. One of the most important metrics: shows the level of profitability or unprofitability of your advertising. We calculate ROI according to the formula: we subtract marketing expenses from the total profit and divide the rest by them. If a this indicator more than 100% - your investments are profitable, if the indicator is less than 100% - investments are unprofitable, sadly.
Having determined the main metrics on the basis of which our analysis of the effectiveness of contextual advertising will be based, we look further.
What are we watching?
Nothing is clear, please show me!
Let's look at a specific example of how the analysis of the effectiveness of contextual advertising helps to achieve better performance and increase efficiency. Suppose we are running a context for a small travel agency. Advertising was launched in Yandex.Direct and 200,000 rubles were spent in a week.
We go to Yandex. Metrica and fix that this week we have attracted 60 new customers. We calculate CPL: 200,000/60 = 3,333 rubles. During this time, we have sold our services for a total of 1,000,000 rubles. Using a simple formula, we calculate ROI: (1,000,000 - 200,000) / 200,000 * 100%
In total, our ROI is 400%. Basically good, but it could be better.
We look at the conversion for the past month and see that users who went to the site by clicking in a special placement are better converted into orders. Based on this, we redistribute the budget for special placement for several campaigns.
All information on statistics can be collected in different systems and visualized, for example, like this:
Or use Excel:
And you can, again, simplify your task and use the end-to-end analytics service. This approach will save a lot of time and guarantee the accuracy of all data. AT personal account CoMagic similar table would look like this:
Contextual advertising works, it works for a month, it works for two. And something does not smell like big money. And a hefty stone appears in front of you, and on it is an inscription: “If you optimize advertising, you will receive a large revenue (or maybe a small one). You will increase the advertising budget - the revenue will increase, and the profit - as God sends.
You can cast lots, trust your intuition, or write to the "Battle of Psychics". I trust analytics and numbers more. I will tell you my shamanic secrets of how to "feel" the effectiveness of contextual advertising.
How to measure the effectiveness of contextual advertising
The key indicator for all advertising channels is the cost of purchase or CPO (Cost Per Order). To calculate the purchase price, you need to divide the acquisition costs by the number of paid orders. CPO can be calculated both for the entire site and for each channel separately.
The lower the purchase price, the more effective the channel, campaign or ad group is performing
So that the number of applications received from each channel does not have to be determined by coffee grounds, I recommend that you first, at least a couple of weeks in advance, set up goals in Yandex.Metrica and Google Analytics.
And in order to understand how much to lay out "dough" in order to receive one order, you need to calculate how much money it brings on average for the business.
It is necessary to take into account the average check, and marginality, and how many percent of the income we are ready to spend on attracting new customers, so that there is still money left for a car for our son and an apartment for our mistress.
It is better, of course, to use the average LTV (Lifetime Value is the profit that we receive from one client for the entire time of cooperation with him). But such statistics have been collected for years, and they are not always available.
Lena, I didn’t understand anything, explain normally!
To make it clearer, I will explain using the example of old business indicators of one of our clients - an agency selling tours in the sanatoriums of the Caucasus and Belokurikha. Website conversion - 2.43%, on average, the agency earned 7,060 rubles from one sale.
Let's define the boundaries of our greed - indicators of the effectiveness of contextual advertising, which we will focus on. In order not to go into debt and loans, you can spend no more than 100% of the income that it theoretically brings to the campaign to attract one lead.
But you have to pay for telephony, pay salaries to employees, and the services of LeadMachine are also not free. And the business owner wants to see profit: the more, the better. With this in mind, we realized that we dream of spending about 20% of the profit on attracting, which is 1,412 rubles per sale or 212 rubles per application (CPOo - the expected sale price and CPLo - the expected lead price).
As a result, we have determined the intervals in which we want to see key indicators.
Expectation and reality
We calculated key indicators: the value we are striving for, and the upper limit from which we must stay away. Now, rather, we open Yandex.Metrika and Google Analytics and consider what the real CPL is equal to.
In Yandex.Metrica, the number of requests for all channels can be viewed in the standard report "Sources, Summary".
In the table that opens, select the goal and see how many applications each channel brought.
Context costs can be viewed in Yandex.Direct and Google AdWords advertising services. We divide the costs by the number of applications and get our real cost per lead. Then we either dance or cry.
If the lead price suits you, in pursuit of profit, just pour more money in advertising budget. But after such manipulations, CPL can also grow up, so you should always keep your finger on the pulse.
If the client is too expensive
If the price of a lead is not satisfactory, or, even worse, it is more than the maximum allowable value, the effectiveness of contextual advertising leaves much to be desired. So you need to dig deeper.
It is better to look at statistics on Yandex.Direct in Metrica. Open the standard reports again, select "Direct, summary" in the sources, specify the desired goal.
Now you know the number of applications for each campaign. Expenses can be viewed in the "Direct - expenses" report, if the integration of the two services is configured correctly, or in "Yandex.Direct".
Google AdWords statistics on the number of requests and costs are best viewed in Google Analytics. To do this, select "Traffic Sources" - "AdWords" - "Campaigns". Previously, two services must be synchronized with each other.
Campaigns where the cost of the application does not suit you can be disabled, checked for errors and optimized.
The effectiveness of contextual advertising in clicks
Another indicator that will help determine the effectiveness of contextual advertising is CPC, or cost per click.
Let's go back to our example. In an ideal world, we spend no more than 212 rubles per application. Website conversion - 2.43%. Those. Approximately every 41st visitor to the site leaves an application. Let's call this indicator n, we will calculate it according to the following formula:
Then the CPC in an ideal world does not exceed 212 / 41 = 5 rubles 17 kopecks. This value can be set in Direct and AdWords, as maximum price clique. But there will be few impressions and clicks, so it is better to gradually increase this parameter until you achieve the desired traffic.
When increasing your maximum cost per click, remember that there is a limit beyond which ruin and poverty can await you. It's easy to calculate:
The effectiveness of contextual advertising in ROI
End-to-end analytics services make all these calculations simple and automated. You understand the costs for each channel and see the revenue, see the return on investment.
If it is not possible to use such a service, simply control the costs and application prices for each channel, compare them with income. How this is done, I tried to explain as simply as possible.
But be prepared that optimizing ads without taking into account ROI for each channel can backfire. In our example, the cost of an order from Yandex.Direct exceeds CPLo by 14%, and orders from Google AdWords by 52%. But AdWords has a higher ROI: we received more profit from this channel for 22 requests than for 23 from Yandex.Direct. So it goes.
Thanks to the optimization of work with contextual advertising, in just one month we managed to halve the monthly budget and quadruple the number of clicks and hits. The activity of sales managers was reduced to work with the incoming flow. What steps have been taken to achieve these results?
Veniamin Bakalinsky,
CEO, iTrex
In this article you will learn
- How to evaluate the optimization result
Contextual advertising is one of the most popular online sales tools: in 2013, it accounted for three-quarters of all advertising budgets. Many people use it, but contextual advertising effectiveness not everyone sees. Typical complaints are: the number of clicks is low, and the cost per click is too high. In addition, many advertisers cite low conversion of clicks into purchases and the difficulty of calculating this parameter among the main problems. Nevertheless, this tool is in demand, because without “context” there is nowhere now. Some try to set up and run everything on their own, since this is not so difficult. Others outsource advertising to agencies. Still others use ad automation systems and robots for holding bids and positions, such as eLama, DirectManager, and so on. However, many leaders commercial services the results are still unsatisfactory.
Best Article of the Month
We interviewed businessmen and found out what modern tactics help to increase the average check and the frequency of purchases regular customers. We have published tips and practical cases in the article.
Also in the article you will find three tools to determine the needs of customers and increase the average check. With these methods, employees always fulfill the plan for upselling.
- Recommendation services: how to use them to increase sales by 60%
According to our estimates, in areas with little competition, there are usually few clicks, in highly competitive areas there are more. But in good positions, the cost per click sometimes rises to completely sky-high values: from $30 to $50 and even up to $90. At the same time, a click on lower positions can cost $5-7.
You should also take into account the conversion of clicks into requests, which is in the range of 1-5%, and the conversion of requests into purchases ranges from 5 to 20%. As a result, the real cost of acquiring a customer is much higher than it might seem. Of course, we should not forget that the quality of the site, both its design and content, greatly affects the conversion into purchases. However, we will leave these questions behind the scenes for now. Let's use the example of our company to consider why setting up contextual advertising is so important and what needs to be done for this.
Why contextual advertising optimization is a need, not a whim
We acted in much the same way as any average advertiser in the market. First, we set up ads ourselves. Then we decided that the agency would manage it more efficiently (especially since most agencies do not take money for running campaigns in Yandex.Direct, but live on a commission from Yandex). Subsequently, the task was transferred to another agency, since the advertising budget began to be spent too quickly, and the effectiveness of contextual advertising remained at the same level. Then there was another agency. Then we transferred the campaign to a personal manager at Yandex. They took it from there and gave it to a freelancer, who was later changed to another. It is hardly worth saying that all the contractors were not accidental: they were recommended by friends, and they all treated us adequately. However, in the b2b sphere, advertising is very expensive, and the profits are by no means fantastic. That is why what the performers gave us did not suit us.
By the time we decided to take PPC optimization back into our own hands, PPC was an important source of customers, but far from the most important. A greater flow was attracted through cold sales and personal recommendations. The budget for the "context" was about 50-60 thousand rubles. per month. We tried to bring it up to 100-120 thousand, but this did not give a visible result.
Advertising costs stopped paying off, and we returned to the optimal budget. "Context" at these costs brought about 15-20% of the company's monthly turnover. In its purest form, it was on the verge of profitability, but we expected that many customers would come again and then their orders would begin to make a profit. This was only partly true, since our main area is b2b, and through contextual advertising, it was mostly individuals who almost never make repeat orders. Therefore, the effectiveness of contextual advertising remained low.
Algorithm for setting up contextual advertising
One of the most serious problems is the exclusion of ad impressions for irrelevant queries of the non-target audience. Like most advertisers, when creating query templates for which our ads will be shown, we actively used negative keywords to cut off unnecessary audiences. However, they did not give the desired result. For example, users can add “in China”, “free of charge” to the phrase “translation agency” – there are a lot of options. Contractors look through requests for statistics from the Yandex Wordstat service and enter everything superfluous into negative keywords. However, negative keywords do not guarantee the complete exclusion of unnecessary audience. We have a request template “translation agency -Ukraine -free -business plan” (negative keywords are indicated in the request with a minus sign - Ed.). An ad on it will not be shown on the request "translation agency in Ukraine", but will be displayed on the request "translation agency in Kyiv" or, say, "translation agency 12345". That is, it is impossible to foresee all the options - the ad will be shown for many irrelevant user requests. CTR decreases and, as a result, the cost of a click in the system increases. So, if such a problem has arisen, all optimization of contextual advertising must be started from scratch.
Step one. We create a semantic core. Analyze all relevant queries according to Wordstat and Google statistics and select from there everything that matches your activity. Take even those requests that are very rare. In terms of meaning, the new semantic core will remain the same.
Then make a table in Excel in which, from the available queries and individual keywords create the maximum number of possible combinations - from simple and short to long and rare: for example, from “contract translation” to “contract translation with of English language into a Russian native speaker. The formulation of combinations must first be approached from the position of logic and knowledge, based on how customers are looking for you. And then - purely "technically": add synonyms and alternatives to all the resulting combinations. For example, if the original request is “translation from Russian into English”, then the languages \u200b\u200bin it are changed to those with which we work, and instead of one request, 50 are obtained. Or all requests with the words “translation agency” are duplicated in the options “translation agency ”, “translation company”, “translator”, “translation center”.
Technically, such requests are created as follows. A query with alternatives is entered into the excel file in the Yandex.Direct format in the following form: “(bureau | agency | center) of translations”, then the file is imported into the campaign in Direct, and then the resulting campaign is immediately exported back from "Direct" in the table. In the resulting file, each request already goes on a separate line. Such an iteration is needed for further processing and generation of requests. We got about 15,000 of them.
All requests must be enclosed in quotation marks - this will indicate to the system that the key phrase must exactly match the request. As a result, it turns out that an ad that matches the key phrase, for example, "contract translation" - is shown only for the request "translation of the contract" and does not appear for the queries "translation of the contract into Russian", "translation of the contract in text view". This will give you an exact match.
Step two. We create ads. Break keyword phrases into thematic groups and write an ad for each group, presumably corresponding to the interests of the one who enters such queries. We got about 1000 ads. Of course, many of them are similar and differ for the most part only in the names of languages, topics or documents.
Step three. We associate with the correct pages on the site.“Putting” all users on the main page of the site is stupid: many will leave if they do not immediately find an answer to their question, and this reduces the effectiveness of contextual advertising, since payment is made for clicks. Therefore, for each ad, you need to specify the appropriate landing page, which would give an answer to the user's request. For some ad groups, the landing page will be the same, but it is important that each page matches the ad in meaning. For example, from the request "translated from English price" the visitor gets to our page with a detailed price list.
Step four. We upload the final information to Yandex.Direct. Adjust the tables to the format that can be uploaded to Yandex.Direct. The description of the format is in the service manual. Pay extra attention to specifying a set of short links that are displayed under the ad text. Here it makes sense to provide links to information that will be of interest to the majority of target users: prices, terms and customer recommendations. Data is uploaded in one file through a special form.
All the work described above (the entire optimization process) can be completed by one specialist in about two weeks, if he treats it with maximum responsibility.
How to evaluate the effectiveness of contextual advertising
Monthly monitoring of the situation showed us very positive changes. average price the click dropped from $4-5 to $0.7 when showing ads in the search engine. In the contextual network (on sites that are partners of the system), it amounted to $0.1. Previously, we did not use context network, because it was rapidly eating up budgets at a click price of $0.5-0.8 and did not give the slightest result.
The click-through rate for all campaigns increased from 0.7-1% to 5-7% on average. We were prevented from getting higher numbers by ads that are shown very rarely (10-20 times a month) and do not generate clicks at all. These are situations when the user is not targeted or our ad did not interest him. We are still working on these announcements. And for individual ads where there are clicks, the CTR can be 10, 20 and even 50%.
- Marketing in a crisis: 3 creative approaches to creating advertising
The fact that a lot of requests and announcements do not work yet (since such requests are entered very rarely) is not scary. There will definitely be a return. Indeed, for rare, low-frequency queries, the user will see your ad, which will be in higher positions relative to other advertisers, or even be the only one. At the same time, a click will cost you many times less than for competitors.
When determining the cost per click, the search engine takes into account the CTR, and this ad has a high CTR. And the higher, the lower the cost per click will be for you. The same query “translation agency” with negative keywords will be significantly more high-frequency and with a lower CTR than a strict query “translation agency”, since in the first case all combinations of the words “translation agency” with any other words, including irrelevant ones, are suitable. This means that the transition to a strict request, as a rule, will cost less.
Automation of bid management. Additionally, you should take care of bid management, because if competitors raise their bids per click for some queries, then your ad may be forced out of impressions and in order to remain in the visibility zone, the price will have to be increased. On the other hand, if competitors' bids are lower than yours, it makes no sense to keep your bids high - you can lower them without losing positions. However, manually tracking them and changing a large number of ads is not an easy task. Therefore, to set up contextual advertising and manage rates, it is better to use auto brokers 1 .
There are many companies on the market now offering similar services for a small fee. subscription fee. We have been using the services of the "Magazion" service for quite a long time. It gave good results, but at some point, the owners, apparently, decided to curtail the project, and the system stopped updating rates. Then we used the DirectManager system, but it did not bring much benefit, so for now we settled on the built-in Yandex.Direct auto-broker. The opinions of advertising experts on this issue differ: some believe that the built-in auto broker “plays along” with the system and sets prices above the optimal ones; others say it works fairly, does everything for free, and updates bets more often than third-party solutions. We haven’t noticed a big difference yet, but we haven’t been very fond of testing: winning 5-7% in bets is not worth the time that needs to be spent on testing. So far, the autobroker is doing its job quite satisfactorily.
1 Autobrokers are automatic CPC optimization systems. They constantly monitor the cost per click and keep it at a minimum level, taking into account the specified campaign parameters. - Ed.
Veniamin Bakalinsky Graduated from the Faculty of Cybernetics of MEPhI in 2002. Completed an MBA course at the State University Higher School of Economics. Prior to organizing the translation agency, he was engaged in marketing and promotion in the companies NeuroK, Ventra and in the TOP magazine.
iTrex Translation Agency founded in 2006. It ranks 26th in the ranking of the largest translation agencies in Russia according to translationrating.ru. Among the clients: AK BARS Bank, SU-155, Mail.Ru Group, Miroglio Group, Ozon.ru, Raytheon Professional Services. Turnover - $ 1.6 million (in 2013). The company is currently preparing to open a branch in Tel Aviv. Staff — 10 employees; over 2500 freelance translators. Official website - www.itrex.ru