How Yandex.Mail support service learned to respond within three hours

    We are trying to make Yandex.Mail so that everything is immediately apparent to it and no help is needed. Alas, this does not always work, so we have a support service. I think that many people in Khabr understand that it is very difficult to organize a support service so that it can quickly respond to all users of such a huge service as Yandex.Mail, and at the same time do not send semi-automated unsubscribes, but essentially help and sort out the problems in the service itself. The last time we talked about how to improve work Yandex.Mail support, reducing the response time to users to one day. And today, when Habrahabr suffered

    all the hubs about managing projects on Megamind, we decided to tell you how we reorganized the support work so that the average response time to people decreased to three hours.

    Do not be afraid of your desires

    More than a year ago, when we just started to change the work of the support service, the average response time was more than three days. Then the answers during the day seemed to us the ultimate dream. But as soon as we achieved this, I wanted more.

    So what about the problem that many online stores and even paid services we use are still not solved (in our subjective opinion), we can handle?

    Problem finding

    In order to understand on the shore what and how affects the response time, we unloaded all the events from our administrative panel for a full month in minutes. Any letter from the user takes time from a support specialist, even if he simply had to be transferred to the next queue. Therefore, it is important to consider not only the arrival of the letter and the answer to it, but also to move to another queue. Closing without an answer takes at least a few seconds - you need to read the letter. We passed this data to analysts and asked them two questions:
    1. Can the support team of the current staff process letters in real time?
    2. What exactly gives the greatest increase in response time?

    Analysts told us the following:
    1. The number of emails from us for a period of time is 96% of the number of incoming messages for the same period. This means that the group copes with the load and could theoretically process most of the tickets “in real time”.
    2. If we consider only tickets closed in less than two days, the average time it takes is 12.3 hours. For different days, this figure ranges from 8 to 16 hours. Despite the fact that there are still 6% of them, during the work on which difficulties arise, we still derived the first metric - the average response time should be reduced by 9.3 hours.
    3. 53% of all letters processed by an employee of a support service during an eight-hour working day arrived on the same day, and 47% remained from yesterday.
    4. The number of tickets opened and closed per day usually varies within 25% in any direction. This difference is also large enough to provoke delays in response time.

    Matter of technology

    The main thing that we understood: the main delay in the response was caused by old tickets accumulated in the queue. So we formulated the first task: to reduce the level of open tickets once, while maintaining the average load. To solve it, we did not even need to change the work schedule or hire new employees. We just organized a couple of subbotniks and sorted out all the hanging problems.
    This change alone gave an increase in the average response rate of 6 hours (12.3 - 6 = 6.6). That is, to our goal in 3 hours we needed to reduce the response time by half.

    It remained to synchronize the work schedule of the support staff, as the floating schedule on weekdays gave us about 3-4 hours to an average time delay.

    As can be seen in the graph, it was necessary to pay attention to the periods from 7 to 11 and from 19 to 22 hours. At the same time, from 13 to 16, on the contrary, you can unload, which we did.

    The new metric was an empty queue at 22-00 daily and a new work schedule. The effect is observed on the chart:

    And the winner is ...

    Only organizational changes allowed us to grow in terms of response more than twice - from 30 to 75 percent of responses within three hours. The number of letters per day grew by only 10 percent, thereby our fear of a significant increase in letters due to quick replies was in vain. To make out the remaining 25%, two shifts must be introduced, since most of them are questions that came up at night Moscow time. Now we answer them within 8-12 hours. But we will talk about this next time.

    We are proactive

    80% of letters coming to the support service are typical questions that have standard answers: the user simply did not find some kind of setting or wants to find out how some kind of service works. The robot can answer such questions.

    To solve this problem, we made a feedback form in the form of a pop-up with a noticeable search bar for questions and answers, a web screenshot and the ability to immediately attach a problematic letter in the form of an attachment.

    In terms of benefits, we were compared with the usual form of feedback. The key metric of the project was the conversion of those who entered the form into those who wrote to the support.

    The resulting numbers are as follows:
    • Conversion in a new form - 23%.
    • Conversion in the old form - 56%.

    From those who wrote to the support service:
    • 12% came from search results (that is, they were not satisfied with our answer);
    • 88% came from the start page (they didn’t look for anything, didn’t click on “Find”, immediately click on “Write to us”).

    Those who did not write:
    • 41% went to Help from the first page;
    • 32% used the search and found the answer (30% of them used the sujest);
    • 27% closed the form immediately or while writing a letter

    The new search form was 33% more effective than the old one.
    It also became clear that people, contrary to popular belief, are able not only to write in support, but also to independently use the search, read and find answers - it is enough to give them a convenient tool.


    Yandex.Mail is a free email service with one of the fastest support services even among paid services. 75% of our users receive answers to their questions within three hours, and 90% within 24 hours.

    We managed not to grow in the number of letters and at the same time powerfully improve the response time by the current team - without any major changes in the work schedule. And we do not stop there. Now we plan to learn how to answer important questions within an hour, start automatically checking the quality of the answer, learn how to calculate the user's karma on the service, in order to more accurately determine the level of his knowledge and skills.

    Also popular now: