Interview with Ivan Kozlov, vice president of mobile products Aviasales.ru
We are starting to publish business interviews with experts from the mobile industry about products, mobile promotion, analytics, user acquisition and business in general.
The first expert who was interviewed by Alexey Pisarevsky was Aviasales Vice-President for Mobile Products and our long-time client and good friend, Ivan Kozlov.
Hello! I want to start unexpectedly. We know a lot about Aviasales, but you also have a second hotel booking product - Hotellook. There is little information about him. Tell me what's wrong with him? What are the differences and what do these two products have in common?
')
We are talking less about Hotellook, since the project itself is currently smaller. Now there is a phase of active growth, for this summer we have grown very much. The fundamental difference from Aviasales is only one: Hotellook - about hotels, and Aviasales - about tickets. Otherwise, everything is similar: this is the same search, with the same problems as in the tickets. But there are several positive differences: in hotels, margins are 3-4 times higher, and this is an international project, as all of our partners: Booking.com, Agoda.com, Hotels.com and others are international players who are able to sell hotels by around the world, not just in Russia. Entering the overseas market with Hotellook is much easier than with Aviasales.
That is, Hotellook is already ready to move abroad?
The product is ready for a long time, and now the share of foreign sales is about 30-35%.
Do you mean overseas buyers or booking overseas hotels?
We are talking about foreign buyers who book hotels through our service.
But I know that you were not actively promoting it abroad? From which countries do you have the most users?
Yes, that's right, almost all of this is done with the help of organic matter: ASO and so on.
Overseas sales Hotellook
Have you planned from the very beginning that your product abroad will be so in demand?
We assumed that a hotel product abroad is easier to promote in terms of business development. Speech about the signing of local players. If you want to be in Brazil, you need local a / k, local agencies. With hotels, everything is much easier.
And you do not suffer from the fact that the brand Hotellook is still poorly recognizable against the backdrop of a terrible worldwide booking? With Aviasales, the story is different, everyone knows this brand in Russia, it is easier to promote. How much does this change the picture for you?
Yes, Booking.com is, of course, a global giant, almost synonymous with the concept of "hotel", and it is extremely difficult to compete with it, given the budgets that are there. But we have some UTP. For example, Booking.com, from the point of view of its pricing policy, is not always a profitable service, and our users understand this, therefore they use our services. As for Aviasales, the brand is really very famous, and, on the one hand, it is easier to promote it, but on the other hand, you yourself know perfectly well that our offer has already been burned out several times, and to find mobile traffic under it in Russia every day harder and harder.
Growth of Hotellook installations for 2016
Yes it's true. And if we talk about marketing metrics, you said quite a lot that in Aviasales it’s hard to count a unit-economy, and the user doesn’t immediately make a purchase after installation. In Hotellook, everything is about the same, or have their own features?
Approximately everything is the same: the time of the first transaction is the same as in air tickets, there are no fundamental differences. From the point of view of geography, everything is the same as in airline tickets: the top search cities in Russia are Moscow, St. Petersburg and Sochi, but there is a nuance - the margins are much higher.
You once said that Aviasales is considered normal for you if you have returned 50% of paid traffic in a year, taking into account all possible losses. In Hotellook about the same?
No, Hotellook is better. It is a little more difficult to count here, since everything is simple in airline tickets: I bought a ticket, a transaction has passed, there is booking and partners pay during the reporting period. In hotels everything is different, there booking does not involve receiving money immediately, as free cancellation is possible, many different nuances, and we get money from partners only after some time, after the guest left the hotel, therefore the economy considered differently. But, in general, what we see on margin, Hotellook's unit economy is better.
Aviasales is installed in every sixth smartphone in Russia
Therefore, since it’s profitable to sell the hotels, you are aggressively trying to sell them to those who buy a plane ticket. How effectively do you do this? Is it easy to drag users to Hotellook? Are many users mad at you for this?
Of course, there are cross-promo campaigns from Aviasales and Hotellook. We send them quite a lot, and this is the highest quality and loyal traffic. Thus, we catch up with foreign audiences. In general, fears from advertising have been and are, but according to the results of our research, a not very annoying and fairly native looking advertisement does not reduce our metrics. Someone just leaves, but, in fact, it does not affect our performance, and everything is quite difficult to count. This is a game for a very long distance.
And how do you generally feel about not completely honest apseilmas that are misused by many services? For example, I did not remove any tick when booking and you have written off the n-th sum of money for insurance.
I take this very negatively, namely, the form of filing. In fact, it is no secret to anyone that air tickets have very low margins, high competition, and price plays the biggest role for the user, so everyone dumps. Someone sells with a minimum margin, someone to zero, someone even in the minus, so in order to somehow survive, you have to use some tricks. I am negative about such tricks, but from the point of view of agency service it is right to try to make money on additional services: insurance, sms-notifications, cancellation without commission, return of the ticket. Just sometimes the form of filing is too aggressive. For our part, we are very actively working on this and we point out to agencies that we don’t think that we should be doing this. If necessary, stop working with them. Now we were able to ensure that at least all the checkboxes were turned off by default, and that the currency was written off in the user's currency.
Probably a lot of questions are being asked to you about the crisis, because it is believed that he hit the travel industry the hardest. Quite a lot of time has passed since the ruble fell, now, looking back, how do you see, where would you be if there were no crisis? Was it hard to overcome?
Of course, we felt the crisis: the demand for our service changed, more domestic flights became available, the share of overseas flights fell. If it were not for the events that occurred, we would now be in a completely different place in finance, turnover, etc., but we very quickly adapted to the new realities.
And who is generally profitable now in the market of sales of airline tickets and hotels, well, of course, except for you?
Provocative question :-). Honestly, I don’t know, apparently, the players who have been on the market for 3-5 years could find some niche in which they can work quite successfully, at least without a loss. For my part, it would not be correct to call a company specifically, but there are definitely profitable players.
Listening to you, it seems that the market is hard. Well, what about travel startups who want to launch a meta-search? Can I earn something?
In the travel-sphere you can run a lot of things. Many uncovered or poorly covered niches. Regarding the metasearch, the train has probably already left here. Now meta-search is expensive, difficult and incomprehensible why. It is unlikely that he will be able to offer what we can not offer. It is also about the hotel meta-search, in which there are both global and international players.
Okay, let's close the market theme and talk a little about more mundane topics. How do you work on increasing retention in your mobile products? For example, push notifications have always been the traditional way to return a user. Is it true that this does not work as well as it used to? Generally, what difficulties do you see?
Push notifications work. The question is "how to ask" or "what to send." As part of Aviasales (there are no push notifications in Hotellook), we carefully ask the user if he wants to receive notifications. We send him information about price changes for those routes that are interesting to him. Push still works, just the conversion from it has never been colossally high. Opening applications is there, de facto Retention is increasing, but there is almost no direct connection with conversion and purchase.
Well, you have a huge number of “sleeping” users who have refused to push, and your cool newsletters do not read, how to deal with them? In the web, large companies use retargeting for this. In the mobile - it is still quite a young niche. What do you think about him?
Of course, we tried mobile retargeting, since with our current install base this thing is absolutely necessary. Something turns out better, something worse, until the main concern is that the best efficiency is obtained at a very small volume, and somehow we do not get out of this story. But we are actively working on it. Now we are considering retargeting on Facebook, we will have a couple of launches in the coming month, and we expect their tools to attract old users.
Cool! We at Mobio really believe in mobile retargeting and in beta with some clients we are already testing our new product GetLoyal , a minute of advertising :-). But back to the questions. In Aviasales, a mobile unit is essentially a separate business unit. You even have an office separately. Quite a lot of companies do this: Yandex and Yandex Taxi, Mail.ru and Maps.me. Does it help you somehow? What are the disadvantages of this approach?
Yes, in our department a model is historically used, as in Mail.ru. We are, indeed, a separate and self-sufficient business unit. It seems to me that now tools and technologies allow communicating at a distance, I don’t see any problems in this.
Tell me about the team: how many people work in the St. Petersburg office, what kind of people are they, what do they do
We employ about 40 people. There are several guys who came from Thailand to St. Petersburg, working in our office, but in other departments. Almost everything is done in-house, there is development, testing, design, prototyping and product managers, and marketing.
And how are the processes arranged? Are there development cycles, weekly meetings?
It all depends on the product manager and his agreement within the team. Some teams work on Scrum, some on Kanban. I really don't care, the main thing is the result. And so, we have all the standard attributes of agile development: planning, retrospective, demo, daily short meetings, everything is there.
Are you personally personally immersed in the process?
I use jira to watch what the guys are doing, what stage they are at, actively participate in the demo and interact with product managers - this is an essential part of my work. Well, it turns out that I hold all mobile marketing in my hands, there is no intermediate link between me and the final implementers.
Orchestra man, share a secret, how do you not bother all this? You have been doing the same product for quite a long time, a lot of routine, the same tasks.
It is difficult to answer this question. Still, I haven't been working for a very long time, only 4 years old :-) We have more than ten products. They are completely different, with different target audiences, there is simply no time to be bored. In fact, in 4 years the mobile industry has changed a lot, there was a completely different kind of market, and it is extremely interesting to participate in its transformation. If I worked on the web, I would probably be bored. This system has long been established, some changes occur, but with little dynamics.
Finally, what do you advise to read, people interested in the industry? What do you read?
I read more or less classic blogs, foreign technical publications, subscribed to the classic newsletters. I have a very well formed ribbon on Facebook, from which I learn a lot of interesting things. I will not tell you anything specific, except, of course, my channel in Telegram.
Thank you for agreeing to chat. We are waiting for Aviasales and Hotellook loud announcements!
The first expert who was interviewed by Alexey Pisarevsky was Aviasales Vice-President for Mobile Products and our long-time client and good friend, Ivan Kozlov.
Hello! I want to start unexpectedly. We know a lot about Aviasales, but you also have a second hotel booking product - Hotellook. There is little information about him. Tell me what's wrong with him? What are the differences and what do these two products have in common?
')
We are talking less about Hotellook, since the project itself is currently smaller. Now there is a phase of active growth, for this summer we have grown very much. The fundamental difference from Aviasales is only one: Hotellook - about hotels, and Aviasales - about tickets. Otherwise, everything is similar: this is the same search, with the same problems as in the tickets. But there are several positive differences: in hotels, margins are 3-4 times higher, and this is an international project, as all of our partners: Booking.com, Agoda.com, Hotels.com and others are international players who are able to sell hotels by around the world, not just in Russia. Entering the overseas market with Hotellook is much easier than with Aviasales.
That is, Hotellook is already ready to move abroad?
The product is ready for a long time, and now the share of foreign sales is about 30-35%.
Do you mean overseas buyers or booking overseas hotels?
We are talking about foreign buyers who book hotels through our service.
But I know that you were not actively promoting it abroad? From which countries do you have the most users?
Yes, that's right, almost all of this is done with the help of organic matter: ASO and so on.
Overseas sales Hotellook
Have you planned from the very beginning that your product abroad will be so in demand?
We assumed that a hotel product abroad is easier to promote in terms of business development. Speech about the signing of local players. If you want to be in Brazil, you need local a / k, local agencies. With hotels, everything is much easier.
And you do not suffer from the fact that the brand Hotellook is still poorly recognizable against the backdrop of a terrible worldwide booking? With Aviasales, the story is different, everyone knows this brand in Russia, it is easier to promote. How much does this change the picture for you?
Yes, Booking.com is, of course, a global giant, almost synonymous with the concept of "hotel", and it is extremely difficult to compete with it, given the budgets that are there. But we have some UTP. For example, Booking.com, from the point of view of its pricing policy, is not always a profitable service, and our users understand this, therefore they use our services. As for Aviasales, the brand is really very famous, and, on the one hand, it is easier to promote it, but on the other hand, you yourself know perfectly well that our offer has already been burned out several times, and to find mobile traffic under it in Russia every day harder and harder.
Growth of Hotellook installations for 2016
Yes it's true. And if we talk about marketing metrics, you said quite a lot that in Aviasales it’s hard to count a unit-economy, and the user doesn’t immediately make a purchase after installation. In Hotellook, everything is about the same, or have their own features?
Approximately everything is the same: the time of the first transaction is the same as in air tickets, there are no fundamental differences. From the point of view of geography, everything is the same as in airline tickets: the top search cities in Russia are Moscow, St. Petersburg and Sochi, but there is a nuance - the margins are much higher.
You once said that Aviasales is considered normal for you if you have returned 50% of paid traffic in a year, taking into account all possible losses. In Hotellook about the same?
No, Hotellook is better. It is a little more difficult to count here, since everything is simple in airline tickets: I bought a ticket, a transaction has passed, there is booking and partners pay during the reporting period. In hotels everything is different, there booking does not involve receiving money immediately, as free cancellation is possible, many different nuances, and we get money from partners only after some time, after the guest left the hotel, therefore the economy considered differently. But, in general, what we see on margin, Hotellook's unit economy is better.
Aviasales is installed in every sixth smartphone in Russia
Therefore, since it’s profitable to sell the hotels, you are aggressively trying to sell them to those who buy a plane ticket. How effectively do you do this? Is it easy to drag users to Hotellook? Are many users mad at you for this?
Of course, there are cross-promo campaigns from Aviasales and Hotellook. We send them quite a lot, and this is the highest quality and loyal traffic. Thus, we catch up with foreign audiences. In general, fears from advertising have been and are, but according to the results of our research, a not very annoying and fairly native looking advertisement does not reduce our metrics. Someone just leaves, but, in fact, it does not affect our performance, and everything is quite difficult to count. This is a game for a very long distance.
And how do you generally feel about not completely honest apseilmas that are misused by many services? For example, I did not remove any tick when booking and you have written off the n-th sum of money for insurance.
I take this very negatively, namely, the form of filing. In fact, it is no secret to anyone that air tickets have very low margins, high competition, and price plays the biggest role for the user, so everyone dumps. Someone sells with a minimum margin, someone to zero, someone even in the minus, so in order to somehow survive, you have to use some tricks. I am negative about such tricks, but from the point of view of agency service it is right to try to make money on additional services: insurance, sms-notifications, cancellation without commission, return of the ticket. Just sometimes the form of filing is too aggressive. For our part, we are very actively working on this and we point out to agencies that we don’t think that we should be doing this. If necessary, stop working with them. Now we were able to ensure that at least all the checkboxes were turned off by default, and that the currency was written off in the user's currency.
Probably a lot of questions are being asked to you about the crisis, because it is believed that he hit the travel industry the hardest. Quite a lot of time has passed since the ruble fell, now, looking back, how do you see, where would you be if there were no crisis? Was it hard to overcome?
Of course, we felt the crisis: the demand for our service changed, more domestic flights became available, the share of overseas flights fell. If it were not for the events that occurred, we would now be in a completely different place in finance, turnover, etc., but we very quickly adapted to the new realities.
And who is generally profitable now in the market of sales of airline tickets and hotels, well, of course, except for you?
Provocative question :-). Honestly, I don’t know, apparently, the players who have been on the market for 3-5 years could find some niche in which they can work quite successfully, at least without a loss. For my part, it would not be correct to call a company specifically, but there are definitely profitable players.
Listening to you, it seems that the market is hard. Well, what about travel startups who want to launch a meta-search? Can I earn something?
In the travel-sphere you can run a lot of things. Many uncovered or poorly covered niches. Regarding the metasearch, the train has probably already left here. Now meta-search is expensive, difficult and incomprehensible why. It is unlikely that he will be able to offer what we can not offer. It is also about the hotel meta-search, in which there are both global and international players.
Okay, let's close the market theme and talk a little about more mundane topics. How do you work on increasing retention in your mobile products? For example, push notifications have always been the traditional way to return a user. Is it true that this does not work as well as it used to? Generally, what difficulties do you see?
Push notifications work. The question is "how to ask" or "what to send." As part of Aviasales (there are no push notifications in Hotellook), we carefully ask the user if he wants to receive notifications. We send him information about price changes for those routes that are interesting to him. Push still works, just the conversion from it has never been colossally high. Opening applications is there, de facto Retention is increasing, but there is almost no direct connection with conversion and purchase.
Well, you have a huge number of “sleeping” users who have refused to push, and your cool newsletters do not read, how to deal with them? In the web, large companies use retargeting for this. In the mobile - it is still quite a young niche. What do you think about him?
Of course, we tried mobile retargeting, since with our current install base this thing is absolutely necessary. Something turns out better, something worse, until the main concern is that the best efficiency is obtained at a very small volume, and somehow we do not get out of this story. But we are actively working on it. Now we are considering retargeting on Facebook, we will have a couple of launches in the coming month, and we expect their tools to attract old users.
Cool! We at Mobio really believe in mobile retargeting and in beta with some clients we are already testing our new product GetLoyal , a minute of advertising :-). But back to the questions. In Aviasales, a mobile unit is essentially a separate business unit. You even have an office separately. Quite a lot of companies do this: Yandex and Yandex Taxi, Mail.ru and Maps.me. Does it help you somehow? What are the disadvantages of this approach?
Yes, in our department a model is historically used, as in Mail.ru. We are, indeed, a separate and self-sufficient business unit. It seems to me that now tools and technologies allow communicating at a distance, I don’t see any problems in this.
Tell me about the team: how many people work in the St. Petersburg office, what kind of people are they, what do they do
We employ about 40 people. There are several guys who came from Thailand to St. Petersburg, working in our office, but in other departments. Almost everything is done in-house, there is development, testing, design, prototyping and product managers, and marketing.
And how are the processes arranged? Are there development cycles, weekly meetings?
It all depends on the product manager and his agreement within the team. Some teams work on Scrum, some on Kanban. I really don't care, the main thing is the result. And so, we have all the standard attributes of agile development: planning, retrospective, demo, daily short meetings, everything is there.
Are you personally personally immersed in the process?
I use jira to watch what the guys are doing, what stage they are at, actively participate in the demo and interact with product managers - this is an essential part of my work. Well, it turns out that I hold all mobile marketing in my hands, there is no intermediate link between me and the final implementers.
Orchestra man, share a secret, how do you not bother all this? You have been doing the same product for quite a long time, a lot of routine, the same tasks.
It is difficult to answer this question. Still, I haven't been working for a very long time, only 4 years old :-) We have more than ten products. They are completely different, with different target audiences, there is simply no time to be bored. In fact, in 4 years the mobile industry has changed a lot, there was a completely different kind of market, and it is extremely interesting to participate in its transformation. If I worked on the web, I would probably be bored. This system has long been established, some changes occur, but with little dynamics.
Finally, what do you advise to read, people interested in the industry? What do you read?
I read more or less classic blogs, foreign technical publications, subscribed to the classic newsletters. I have a very well formed ribbon on Facebook, from which I learn a lot of interesting things. I will not tell you anything specific, except, of course, my channel in Telegram.
Thank you for agreeing to chat. We are waiting for Aviasales and Hotellook loud announcements!
Source: https://habr.com/ru/post/314370/
All Articles