April 20, 2018 at 15:03Keyword services on the App Store: comparative
Those who are engaged in ASO on an ongoing basis, and even, God forbid, according to different localizations, know: this is not a matter in which you can act on a hunch. Of course, there are a lot of words in the dictionary, but there are also enough products on the market, and serious competition is unfolding for the simplest, associative keys. To maintain a balance between the popularity of the request and the probability of breaking through to the first positions, a thorough analysis of the situation in the corresponding segment is needed.
Today, such systems are engaged in such analysis, which collect data from markets and, processing them, give an estimate of the effectiveness of a key in a numerical value. There are many of them, and in recent weeks our team has a need to audit the available options and select the most promising ones. Under the cut you will find our ten-tool experience for ASO.
Let's start with a short background. We are not the first year on the App Store and, accordingly, the idea of “breaking keys” came to us not for the first time. Initially, for all our ASO needs, we used the well-known Sensor Tower among the masses and still have not seen a reason to refuse it.
This is what the standard Sensor Tower screen looks like. As you can see, for each key, the system displays the following information: Traffic (an estimate of the number of users sending a request), Difficulty (an estimate of the popularity of a request among other developers on the market), the number of applications that appear on request, and the position of your application in the rating. In our opinion, this is quite enough to assess the chances of new keys and the effectiveness of old ones. On the Keyword Spy tab, you can see how competitors are doing, and on the Keyword Suggestions tab, you can see the options generated by the system. Thus, the user gets everything that is needed for the full optimization cycle: he can make a preliminary list of keys, check their indicators, select the most promising ones and, later, see if the application is rising in the issuance.
And everything would be fine, but we are increasingly faced with technical problems. The first calls started back in the days when we used only free accounts - then this could be attributed to the fact that we want too much from the default truncated functionality, especially since problems appeared only under heavy load. But even after switching to a paid tariff, Sensor Tower soon began to issue old errors with increasing regularity. Multi-component phrases were difficult, often had to enter them several times. At times, the system simply “lay down” and for some time did not respond to incentives in principle. And finally, in some localizations, the calculation was suspicious: the same values could be repeated over and over for a series of keys.
Communication with the administration did not correct the situation, and we understood: it was time to see what else the market offers. Based on our experience with Sensor Tower, we searched for services that could offer the following:
Based on reviews, recommendations, as well as the availability of a demo version or “free of charge” tariff plan, we selected nine services to work with them ourselves and check both for compliance with our requirements and for basic convenience. We entered the feedback on the listed criteria in a special table - at the end of the experiment, it acquired the following form:
[ full size ]
Already from this picture it is approximately clear which options are more suitable for us and which are less. But I would like to talk a little more about each of the tools and the impression he left.
Appannie
The idea to pass optimization into the hands of AppAnnie, of course, is a tempting one - the resource is very solid, their reports were read and used in their strategies, probably all. The possibilities of a free account are greatly reduced, but a page with premium benefits promises everything you can wish for: traffic, competitiveness and comparison of requests to select the most effective ones. The Keyword Search tab also catches your eye with blocked useful indicators like Estimated Downloads (estimated number of downloads). In general, the system is very similar to SensorTower, and the brand allows you to count on high quality service.
One thing is bad - the prices are appropriate. Following negotiations with managers, AppAnnie topped our personal ranking of the most expensive tools. Accordingly, it had to be postponed until better times.
Searchman
We resorted to the services of this service earlier, when there were no complaints to the Sensor Tower yet - it is a lifesaver for cases when imagination or vocabulary is running out. SearchMan shamelessly conveys to competitors against each other, collecting data about the keys used and creating a common semantic field for applications of the same type, which can be accessed on the Keyword Library tab. The service also selects competitors for you and conducts a direct comparison of search queries and search results. In general, there are many opportunities to look at people and fit as nicely as possible into a niche.
The rest of the SearchMan functionality, for the most part, corresponds to the usual set for us (traffic, competition, place in the ranking). Moreover, the service further facilitates the optimizer’s task by correlating traffic and competitiveness indicators in a specific KEI indicator. But there is one unexpected omission: his list of countries is more than modest. Even the most expensive packages promise no more than twenty, and those who do not intend to pay a lot will have to be content with three - the United States, Britain and Japan. Unfortunately, for us this is a key moment, and I had to abandon the idea of making this really rich and convenient tool the main one. But as an additional one, it will also serve us - the selection and comparison of keys here can be carried out even on a free account.
AppRankCorner
A rare completely free copy on our list. Against the background of competitors' prices, such generosity raises suspicions, and, sadly admit, not baseless. In theory, AppRankCorner displays all the main parameters (traffic analogues, chances to break into the top, the number of applications issued on request) and even offers a bonus in the form of recommended keys. In fact, the service either does not work at all, or it works according to the mood - we still could not find a single word whose indicators would be non-zero. There are no complaints about the recommendations, but joy from them alone is not enough.
Applyzer
AppLyzer evokes mixed feelings. On the one hand, the site is convenient, the interface is extremely well thought out - working with it is a pleasure. Price categories are quite fractional; unlike other services, where, as a rule, two or three basic tariffs are presented, it is easier to find the optimal point between the price / volume axes of services. In the paid version, the possibilities for comparing with competitive products are very tempting, the imagination draws something like a more compact SearchMan. In general, everything is for people ... except for commonplace indicators of traffic and competitiveness, which simply do not exist. The service provides only and exclusively information about the position in the ratings of different countries and categories (with dynamics, graphs and other valuable insights). This approach to data analysis also has the right to exist,
Appfollow
A concise and simple service with fairly limited functionality, which again comes down to tracking application positions for certain requests. It is proposed to evaluate the effectiveness of options by an atypical set of parameters: next to the number of competitive products and the current position in the list, the presence of a keyword in the heading and subheading is indicated. This, of course, has its own logic - it really affects the ranking - but still, such things are easier to verify manually than, say, competition. The site also has a tab called Comparison, which, in theory, allows you to compare yourself with competitors, but the comparison is made to a very limited circle of criteria lying on the surface (size, price, reviews and so on).
At the same time, everything related to scaling and pricing is implemented very reasonably. Here we also have an economical system that allows you to “collect” the necessary services and limits by adjusting the total amount. So, our team, which loves a geographic scope and works on several projects at once, would buy around five and a half thousand a month, and some small studio with one main project and a verified list of foreign markets could easily fit in one and a half .
Appradar
The main disadvantage of the service is its persistent desire to break into your account in the store and merge with it; It is comforting, however, that most of the functions related precisely to optimization for markets can be used without it. In all other respects, Appradar makes a good impression: the screen is successfully divided into zones, so the list of queries is easy to edit, and the recommendations serve as unobtrusive hints. No less sensibly organized and a tab with competitors whose keys are displayed in the format of a neat comparative plate. The rating system, however, does not fully correspond to our usual one - in addition to the inevitable positions in the ranking and the number of applications, there is only one column here - Popularity. But such a calculation is better than none at all.
And now about the sad, that is, about the price tags. Appradar places very stringent restrictions on both the list of countries and the number of applications and keys for each. The cheapest tariff (cheapness, it must be said, is very relative) allows only two products and five localizations, and in order to be able to process at least five applications at a time, you will have to pay the maximum. By the way, the limit is set on the number of competitors that can be monitored, but quotas are much more generous here. In other words, this is an option for those who care about quality and quantity that is not too fundamental - unfortunately, we needed both.
Assodesk
A very rich, but also very erratic set of functions - it is all the more difficult to investigate because the part is blocked for “trial” users. Each key is evaluated in terms of not only the popularity and number of associated applications, but also its potential in Search Ads. You can get information about competitors on three different tabs, some of which contain additional information that is not presented on the main screen - for example, the local analogue of Difficulty. There are also key offers available by click, but they are issued in small portions and in a separate window, which slows down the work. In a word, there is a lot of everything, but it is difficult to say whether it will be convenient to work with the service on an ongoing basis. We were also confused by some zero indicators - they did not prevail, as in AppRankCorner, but did not coincide with what other services showed.
In service packages there is no separation by the number of localizations, which was pleasantly surprised. The main difference is in the number of slots for products and keys, as well as not too critical additions to the functionality. However, according to the FAQ, reasonable price plans are only suitable for those who regularly update one product - this, coupled with the shortcomings listed above, approved us in the decision to hold back the money.
Mobile action
The indicators practically duplicate Sensortower, with minor, for the most part nominal exceptions - this immediately put us to the tool. In the ASO block, all the points that I would like to see are clearly highlighted: evaluation, tracking, recommendations and competitors, plus an addition in the form of a translator for localizations (I could not test the quality of the translation in the demo version, however). Quantitative restrictions are humane, even in the lowest price category; countries are practically in full force.
This service met all the requirements that we listed at the beginning of the article, so we tested it with special passion. In general, the team was very pleased: the count is done quickly and without glitches, both for words and phrases. Employees noted the quality selection of competitors and convenient sorting of recommendations. Advantages of the interface and control were also revealed: recommendations were successfully built into the Research window, the ability to delete the entire list of keys at once was very popular. An unexpected minus is that the service is very chatty, constantly pop-up dialog windows quickly get on your nerves.
Apptweak
With Apptweak, we were lucky to arrange a short trial period with access to a full range of functions - without any irony, we were lucky, the service turned out to be solid. The indicators, again, correlate well with those offered by the Sensor Tower, and the correctness of the calculation does not cause concern. An additional KEI parameter, which we have already seen with SearchMan, also appears here. There are several tens of localizations; this seemed sufficient even to us; there are no restrictions on the number of applications, on the number of keys - the reserve is quite decent. In general, the initial inspection of Apptweak was no problem.
Then the grounded test began, with real keys and the work of the entire ASO team. The feedback that was collected after a week of testing was more mixed than in the case of Mobile Action - there were negative reviews and praise. In general, the tool seemed slightly overloaded and not too intuitive. It lacks compactness: it is often shared that it would be more convenient to have on one screen (for example, localizations or competitors). There are, however, successful solutions - the already mentioned option of quick deletion, unloading, counting characters in a string with keys. As for the main functionality, the quality of the assessment was considered acceptable, although not without failures, especially with regard to the Rank parameter and multi-component phrases.
Summarizing:
In general, the solutions presented on the market are diverse. We evaluated them based on our own preferences, but tried to highlight various aspects, so that people with other priorities could also take note of something. We will be glad to hear about your experience with these or other services and add to the list of recommendations.
Today, such systems are engaged in such analysis, which collect data from markets and, processing them, give an estimate of the effectiveness of a key in a numerical value. There are many of them, and in recent weeks our team has a need to audit the available options and select the most promising ones. Under the cut you will find our ten-tool experience for ASO.
Let's start with a short background. We are not the first year on the App Store and, accordingly, the idea of “breaking keys” came to us not for the first time. Initially, for all our ASO needs, we used the well-known Sensor Tower among the masses and still have not seen a reason to refuse it.
This is what the standard Sensor Tower screen looks like. As you can see, for each key, the system displays the following information: Traffic (an estimate of the number of users sending a request), Difficulty (an estimate of the popularity of a request among other developers on the market), the number of applications that appear on request, and the position of your application in the rating. In our opinion, this is quite enough to assess the chances of new keys and the effectiveness of old ones. On the Keyword Spy tab, you can see how competitors are doing, and on the Keyword Suggestions tab, you can see the options generated by the system. Thus, the user gets everything that is needed for the full optimization cycle: he can make a preliminary list of keys, check their indicators, select the most promising ones and, later, see if the application is rising in the issuance.
And everything would be fine, but we are increasingly faced with technical problems. The first calls started back in the days when we used only free accounts - then this could be attributed to the fact that we want too much from the default truncated functionality, especially since problems appeared only under heavy load. But even after switching to a paid tariff, Sensor Tower soon began to issue old errors with increasing regularity. Multi-component phrases were difficult, often had to enter them several times. At times, the system simply “lay down” and for some time did not respond to incentives in principle. And finally, in some localizations, the calculation was suspicious: the same values could be repeated over and over for a series of keys.
Communication with the administration did not correct the situation, and we understood: it was time to see what else the market offers. Based on our experience with Sensor Tower, we searched for services that could offer the following:
- Clear and efficient key evaluation system
- A wide range of localizations - we work with different countries
- Tips from a recommender system
- The ability to select keys with an eye on competitors: their requests, positions in the ranking
- The ability to consider a large number of keys (ideally, several hundred) for several applications at once (ideally, a dozen)
- Not too sky-high price
Based on reviews, recommendations, as well as the availability of a demo version or “free of charge” tariff plan, we selected nine services to work with them ourselves and check both for compliance with our requirements and for basic convenience. We entered the feedback on the listed criteria in a special table - at the end of the experiment, it acquired the following form:
[ full size ]
Already from this picture it is approximately clear which options are more suitable for us and which are less. But I would like to talk a little more about each of the tools and the impression he left.
Appannie
The idea to pass optimization into the hands of AppAnnie, of course, is a tempting one - the resource is very solid, their reports were read and used in their strategies, probably all. The possibilities of a free account are greatly reduced, but a page with premium benefits promises everything you can wish for: traffic, competitiveness and comparison of requests to select the most effective ones. The Keyword Search tab also catches your eye with blocked useful indicators like Estimated Downloads (estimated number of downloads). In general, the system is very similar to SensorTower, and the brand allows you to count on high quality service.
One thing is bad - the prices are appropriate. Following negotiations with managers, AppAnnie topped our personal ranking of the most expensive tools. Accordingly, it had to be postponed until better times.
Searchman
We resorted to the services of this service earlier, when there were no complaints to the Sensor Tower yet - it is a lifesaver for cases when imagination or vocabulary is running out. SearchMan shamelessly conveys to competitors against each other, collecting data about the keys used and creating a common semantic field for applications of the same type, which can be accessed on the Keyword Library tab. The service also selects competitors for you and conducts a direct comparison of search queries and search results. In general, there are many opportunities to look at people and fit as nicely as possible into a niche.
The rest of the SearchMan functionality, for the most part, corresponds to the usual set for us (traffic, competition, place in the ranking). Moreover, the service further facilitates the optimizer’s task by correlating traffic and competitiveness indicators in a specific KEI indicator. But there is one unexpected omission: his list of countries is more than modest. Even the most expensive packages promise no more than twenty, and those who do not intend to pay a lot will have to be content with three - the United States, Britain and Japan. Unfortunately, for us this is a key moment, and I had to abandon the idea of making this really rich and convenient tool the main one. But as an additional one, it will also serve us - the selection and comparison of keys here can be carried out even on a free account.
AppRankCorner
A rare completely free copy on our list. Against the background of competitors' prices, such generosity raises suspicions, and, sadly admit, not baseless. In theory, AppRankCorner displays all the main parameters (traffic analogues, chances to break into the top, the number of applications issued on request) and even offers a bonus in the form of recommended keys. In fact, the service either does not work at all, or it works according to the mood - we still could not find a single word whose indicators would be non-zero. There are no complaints about the recommendations, but joy from them alone is not enough.
Applyzer
AppLyzer evokes mixed feelings. On the one hand, the site is convenient, the interface is extremely well thought out - working with it is a pleasure. Price categories are quite fractional; unlike other services, where, as a rule, two or three basic tariffs are presented, it is easier to find the optimal point between the price / volume axes of services. In the paid version, the possibilities for comparing with competitive products are very tempting, the imagination draws something like a more compact SearchMan. In general, everything is for people ... except for commonplace indicators of traffic and competitiveness, which simply do not exist. The service provides only and exclusively information about the position in the ratings of different countries and categories (with dynamics, graphs and other valuable insights). This approach to data analysis also has the right to exist,
Appfollow
A concise and simple service with fairly limited functionality, which again comes down to tracking application positions for certain requests. It is proposed to evaluate the effectiveness of options by an atypical set of parameters: next to the number of competitive products and the current position in the list, the presence of a keyword in the heading and subheading is indicated. This, of course, has its own logic - it really affects the ranking - but still, such things are easier to verify manually than, say, competition. The site also has a tab called Comparison, which, in theory, allows you to compare yourself with competitors, but the comparison is made to a very limited circle of criteria lying on the surface (size, price, reviews and so on).
At the same time, everything related to scaling and pricing is implemented very reasonably. Here we also have an economical system that allows you to “collect” the necessary services and limits by adjusting the total amount. So, our team, which loves a geographic scope and works on several projects at once, would buy around five and a half thousand a month, and some small studio with one main project and a verified list of foreign markets could easily fit in one and a half .
Appradar
The main disadvantage of the service is its persistent desire to break into your account in the store and merge with it; It is comforting, however, that most of the functions related precisely to optimization for markets can be used without it. In all other respects, Appradar makes a good impression: the screen is successfully divided into zones, so the list of queries is easy to edit, and the recommendations serve as unobtrusive hints. No less sensibly organized and a tab with competitors whose keys are displayed in the format of a neat comparative plate. The rating system, however, does not fully correspond to our usual one - in addition to the inevitable positions in the ranking and the number of applications, there is only one column here - Popularity. But such a calculation is better than none at all.
And now about the sad, that is, about the price tags. Appradar places very stringent restrictions on both the list of countries and the number of applications and keys for each. The cheapest tariff (cheapness, it must be said, is very relative) allows only two products and five localizations, and in order to be able to process at least five applications at a time, you will have to pay the maximum. By the way, the limit is set on the number of competitors that can be monitored, but quotas are much more generous here. In other words, this is an option for those who care about quality and quantity that is not too fundamental - unfortunately, we needed both.
Assodesk
A very rich, but also very erratic set of functions - it is all the more difficult to investigate because the part is blocked for “trial” users. Each key is evaluated in terms of not only the popularity and number of associated applications, but also its potential in Search Ads. You can get information about competitors on three different tabs, some of which contain additional information that is not presented on the main screen - for example, the local analogue of Difficulty. There are also key offers available by click, but they are issued in small portions and in a separate window, which slows down the work. In a word, there is a lot of everything, but it is difficult to say whether it will be convenient to work with the service on an ongoing basis. We were also confused by some zero indicators - they did not prevail, as in AppRankCorner, but did not coincide with what other services showed.
In service packages there is no separation by the number of localizations, which was pleasantly surprised. The main difference is in the number of slots for products and keys, as well as not too critical additions to the functionality. However, according to the FAQ, reasonable price plans are only suitable for those who regularly update one product - this, coupled with the shortcomings listed above, approved us in the decision to hold back the money.
Mobile action
The indicators practically duplicate Sensortower, with minor, for the most part nominal exceptions - this immediately put us to the tool. In the ASO block, all the points that I would like to see are clearly highlighted: evaluation, tracking, recommendations and competitors, plus an addition in the form of a translator for localizations (I could not test the quality of the translation in the demo version, however). Quantitative restrictions are humane, even in the lowest price category; countries are practically in full force.
This service met all the requirements that we listed at the beginning of the article, so we tested it with special passion. In general, the team was very pleased: the count is done quickly and without glitches, both for words and phrases. Employees noted the quality selection of competitors and convenient sorting of recommendations. Advantages of the interface and control were also revealed: recommendations were successfully built into the Research window, the ability to delete the entire list of keys at once was very popular. An unexpected minus is that the service is very chatty, constantly pop-up dialog windows quickly get on your nerves.
Apptweak
With Apptweak, we were lucky to arrange a short trial period with access to a full range of functions - without any irony, we were lucky, the service turned out to be solid. The indicators, again, correlate well with those offered by the Sensor Tower, and the correctness of the calculation does not cause concern. An additional KEI parameter, which we have already seen with SearchMan, also appears here. There are several tens of localizations; this seemed sufficient even to us; there are no restrictions on the number of applications, on the number of keys - the reserve is quite decent. In general, the initial inspection of Apptweak was no problem.
Then the grounded test began, with real keys and the work of the entire ASO team. The feedback that was collected after a week of testing was more mixed than in the case of Mobile Action - there were negative reviews and praise. In general, the tool seemed slightly overloaded and not too intuitive. It lacks compactness: it is often shared that it would be more convenient to have on one screen (for example, localizations or competitors). There are, however, successful solutions - the already mentioned option of quick deletion, unloading, counting characters in a string with keys. As for the main functionality, the quality of the assessment was considered acceptable, although not without failures, especially with regard to the Rank parameter and multi-component phrases.
Summarizing:
- The most promising of the considered services was Mobile Action and Apptweak: both combine an effective assessment scheme, competitor analysis and an extensive list of localizations with adequate rates and a good level of usability. Those who, like us, find the system adopted by Sensor Tower efficient, are likely to be most comfortable working with these sites.
- Appfollow and Applyzer offer an alternative approach, which is based on monitoring the position in the ratings and leaves the work of analyzing this information to the conscience of the optimizer. We prefer services where specially trained algorithms do this, but if you are satisfied with such a scheme, both tools implement it at a good level.
- We did not find reliable completely free solutions, but we were convinced that some free accounts on paid services quite adequately cope with basic operations for a small circle of applications. These include Searchman, Mobile Action and, to a lesser extent, Asodesk.
- The ability to draw ideas from competitors or a recommendation system is, in our opinion, a significant advantage. According to our employees, Searchman, Mobile Action and Apptweak showed the best results in this regard. If we talk not about selection, but about comparing positions by key, Appradar differs in a particularly convenient presentation format.
In general, the solutions presented on the market are diverse. We evaluated them based on our own preferences, but tried to highlight various aspects, so that people with other priorities could also take note of something. We will be glad to hear about your experience with these or other services and add to the list of recommendations.