May 10, 2018 at 11:31How to collect feedback from users? Epic growth story with Ivan Zamesin
The Epic Growth Story is a flash mob from Gett and the Epic Growth Conference, where product, marketing, and design experts share epic growth stories. In the format of a product Advisor blitz interview, Ivan Zamesin shared within the framework of the Epic Growth Story project how to quickly test hypotheses and collect feedback from users using Customer Development.
Watch the interview and read the notes under the cut.
Growth hack in the product?
When I was responsible for the Yandex.Pictures interface, in 2013 we still had a page-by-page interface. The interface looked sad, bad and hard. We rolled out the UX interface and tried to do many different experiments. One experiment is to search for pictures. I came up with the idea to change the frame color of the picture from white to black and text color, respectively, too. We implemented this change in 5 minutes and received a plus of 10% to our metric - the number of viewed images per session. For such an inert thing, like a search with 5 million DAUs, this is a multiple increase.
Growth story that inspires?
Rather, I am inspired by such processes, when all the guys get drunk on one metric, start the process very quickly, hypotheses are promptly cut to the minimum possible and checked by custom, prototypes, inexpensive experiments and the cheapest solutions. First, one hypothesis is checked per month, then one and a half, two, three, four. Inspires when a team reaches that pace. This is the process I bring to all the businesses that I advise.
How to test a hypothesis?
Very often product teams think of features. The hypothesis is equal to a feature that solves the user's need. In the Job Story phase of hypothesis testing, half of these hypotheses die. Job Story provides an understanding of whether to implement this feature or not. And most importantly, there is no need to invest in development and spend a lot of money on testing such experiments.
When the Job Story is confirmed, the next in speed and simplicity is to ask the user to take a socially expensive action, for example: “Write a post on Facebook about how soon such a thing will happen,” “Send it to your friends on Instagram Direct” and so on. The user must confirm by his actions that this is really important. We went through a confirmation, drew a layout, showed it to three people, again asked for confirmation. Okay, it was confirmed, they made a decision on the knee, they got the money.
At some stages, more detail is needed, some stages are simply skipped. There are complex things that the user simply can’t tell, for example, a referral system. The referral system is hard to verify with a custom. Here, rather, you need to come up with the simplest MVP, run on scripts, lend users who use the interface, and look at the result.
Cool Customer Development?
The coolest custdev is the one you do, the bad custdev - which you don't. CustDev is like a sport. Everyone knows that you need to play sports, pump the size of the heart so that the walls thin out, and you do not die from myocardial infarction, but no one does. Same thing with CustDev. Everyone knows that you need to talk with users, but there are always excuses for not doing this.
Do 4 user interviews per week, and you are handsome.
How to collect feedback from users?
The coolest feedback is when the user loads it in tons and you don’t have to do anything for this. In all the teams, in all the businesses that I am currently consulting, I set up processes when the feedback arrives in a push. Chatiks with users, for example. It really works because you get feedback instantly on any matter of interest. If you need to check the layouts, drop the layout into the chat room with the question: “What do you see on this layout? What can you do on this layout? ”And specific questions about functionality. And also ask to evaluate on a ten-point scale how this solves his need.
It is also useful to collect feedback during the sales process. After each sales session, sales managers must write down the reasons why the customer bought or why not bought. It is important to share these points in general chats. Let them automatically fly to the Slack product, designer, team leader, developer, leader. We talked with people, make a script so that with one click you can send all the insiders after the interview to an internal email. In this way, the feedback will circulate constantly.
How to be friends of a designer and a product?
Once I watched a lecture, which greatly changed my attitude towards working with designers. The bottom line is: a huge amount of value is embedded in the work of another person. The main thing here is to remove your ego, understand the other person’s point of view, and ask him a maximum of questions to determine why he formulated the task, why he sent such a model, and so on. You do not have a task to convince another person, your task is to grow a metric. Be sure to generate ideas together, formulate tasks in terms of Job Story. The designer must know, at least, about what the product manager talked with people, then he will make each micro-decision a little better.
Where does product day start?
I always start with stand up, because stand up is a cool topic to load the context of the previous day of the team and adjust the context of the future. Also, the product must have a lot of time to think. Often I found myself in such a state when all day in meetings or in slots I had to “work”, and there wasn’t even half an hour to sit thinking: “Okay, what about the metric?”
What book do you recommend reading?
Necessarily all books of Intercom'a. Intercom - just frenzied devils, they cool themselves focus and broadcast to the outside world. Books "Jobs to Be Done", "on Product Management", "on Starting Up". Necessarily “StartUp”, “Lean UX”, “Pirate Metrics”, “Inspired”, “Ask mom” - this is the basis. I also really liked the book The Subtle Art of Pofigism, in the original The Subtle Art of Not Giving a Fuck. In order to succeed, you inevitably have to make a huge number of mistakes. If you want, you don’t want to, you just need to make a large number of mistakes, and then choose such a sphere and such errors that you would not be very painful from them, but ideally, to enjoy.
Applications on your phone?
The email client Spark, Notion, Airtable, Rocket and Sberbank, Point, Telegram, I madly love Clear - the best Todo due to the fact that it is possible to lock on 15% of the screen. This is the coolest thing that has happened in my application performance.
Cool example of a UX interface?
Airtable is a space thing, I take off my hat to the guys, they solved a difficult task with many levels of abstractions. This is a database that you can show very variably: on a calendar, with a sign. More scripts can be added. Many functions and all of them are accessible and easily perceived by the user.
I see a huge growth point in self-employment. Yes, it’s uncomfortable when you know that there is no guaranteed income, and tomorrow you can’t earn anything. This puts in conditions in which you grow multiple times faster.
There is growth when you choose a more interesting project, in this project you fight for a more interesting responsibility, achieve a percentage increase in metrics - you are promoted and you grow by percent or tens of percent.
And there is growth when you take a risk - with a high probability you fail, with some probability you grow. And the trick of this growth is that you make many times more mistakes, as in the book “The Subtle Art of Pofigism”, and you learn many times faster. The stakes are higher and the result is also multiple higher.
Ivan took part in the Epic Growth Story project, a flash mob from Gett and the Epic Growth Conference, where product and marketing experts share epic growth stories and growls. A selection of the best stories: egconf.ru/growthstory and in the Telegram channel t.me/epicgrowth .
Watch the interview and read the notes under the cut.
Growth hack in the product?
When I was responsible for the Yandex.Pictures interface, in 2013 we still had a page-by-page interface. The interface looked sad, bad and hard. We rolled out the UX interface and tried to do many different experiments. One experiment is to search for pictures. I came up with the idea to change the frame color of the picture from white to black and text color, respectively, too. We implemented this change in 5 minutes and received a plus of 10% to our metric - the number of viewed images per session. For such an inert thing, like a search with 5 million DAUs, this is a multiple increase.
Growth story that inspires?
Rather, I am inspired by such processes, when all the guys get drunk on one metric, start the process very quickly, hypotheses are promptly cut to the minimum possible and checked by custom, prototypes, inexpensive experiments and the cheapest solutions. First, one hypothesis is checked per month, then one and a half, two, three, four. Inspires when a team reaches that pace. This is the process I bring to all the businesses that I advise.
How to test a hypothesis?
Very often product teams think of features. The hypothesis is equal to a feature that solves the user's need. In the Job Story phase of hypothesis testing, half of these hypotheses die. Job Story provides an understanding of whether to implement this feature or not. And most importantly, there is no need to invest in development and spend a lot of money on testing such experiments.
When the Job Story is confirmed, the next in speed and simplicity is to ask the user to take a socially expensive action, for example: “Write a post on Facebook about how soon such a thing will happen,” “Send it to your friends on Instagram Direct” and so on. The user must confirm by his actions that this is really important. We went through a confirmation, drew a layout, showed it to three people, again asked for confirmation. Okay, it was confirmed, they made a decision on the knee, they got the money.
At some stages, more detail is needed, some stages are simply skipped. There are complex things that the user simply can’t tell, for example, a referral system. The referral system is hard to verify with a custom. Here, rather, you need to come up with the simplest MVP, run on scripts, lend users who use the interface, and look at the result.
Cool Customer Development?
The coolest custdev is the one you do, the bad custdev - which you don't. CustDev is like a sport. Everyone knows that you need to play sports, pump the size of the heart so that the walls thin out, and you do not die from myocardial infarction, but no one does. Same thing with CustDev. Everyone knows that you need to talk with users, but there are always excuses for not doing this.
Do 4 user interviews per week, and you are handsome.
How to collect feedback from users?
The coolest feedback is when the user loads it in tons and you don’t have to do anything for this. In all the teams, in all the businesses that I am currently consulting, I set up processes when the feedback arrives in a push. Chatiks with users, for example. It really works because you get feedback instantly on any matter of interest. If you need to check the layouts, drop the layout into the chat room with the question: “What do you see on this layout? What can you do on this layout? ”And specific questions about functionality. And also ask to evaluate on a ten-point scale how this solves his need.
It is also useful to collect feedback during the sales process. After each sales session, sales managers must write down the reasons why the customer bought or why not bought. It is important to share these points in general chats. Let them automatically fly to the Slack product, designer, team leader, developer, leader. We talked with people, make a script so that with one click you can send all the insiders after the interview to an internal email. In this way, the feedback will circulate constantly.
How to be friends of a designer and a product?
Once I watched a lecture, which greatly changed my attitude towards working with designers. The bottom line is: a huge amount of value is embedded in the work of another person. The main thing here is to remove your ego, understand the other person’s point of view, and ask him a maximum of questions to determine why he formulated the task, why he sent such a model, and so on. You do not have a task to convince another person, your task is to grow a metric. Be sure to generate ideas together, formulate tasks in terms of Job Story. The designer must know, at least, about what the product manager talked with people, then he will make each micro-decision a little better.
Where does product day start?
I always start with stand up, because stand up is a cool topic to load the context of the previous day of the team and adjust the context of the future. Also, the product must have a lot of time to think. Often I found myself in such a state when all day in meetings or in slots I had to “work”, and there wasn’t even half an hour to sit thinking: “Okay, what about the metric?”
What book do you recommend reading?
Necessarily all books of Intercom'a. Intercom - just frenzied devils, they cool themselves focus and broadcast to the outside world. Books "Jobs to Be Done", "on Product Management", "on Starting Up". Necessarily “StartUp”, “Lean UX”, “Pirate Metrics”, “Inspired”, “Ask mom” - this is the basis. I also really liked the book The Subtle Art of Pofigism, in the original The Subtle Art of Not Giving a Fuck. In order to succeed, you inevitably have to make a huge number of mistakes. If you want, you don’t want to, you just need to make a large number of mistakes, and then choose such a sphere and such errors that you would not be very painful from them, but ideally, to enjoy.
Applications on your phone?
The email client Spark, Notion, Airtable, Rocket and Sberbank, Point, Telegram, I madly love Clear - the best Todo due to the fact that it is possible to lock on 15% of the screen. This is the coolest thing that has happened in my application performance.
Cool example of a UX interface?
Airtable is a space thing, I take off my hat to the guys, they solved a difficult task with many levels of abstractions. This is a database that you can show very variably: on a calendar, with a sign. More scripts can be added. Many functions and all of them are accessible and easily perceived by the user.
Why an infobusiness and not a product?
I see a huge growth point in self-employment. Yes, it’s uncomfortable when you know that there is no guaranteed income, and tomorrow you can’t earn anything. This puts in conditions in which you grow multiple times faster.
There is growth when you choose a more interesting project, in this project you fight for a more interesting responsibility, achieve a percentage increase in metrics - you are promoted and you grow by percent or tens of percent.
And there is growth when you take a risk - with a high probability you fail, with some probability you grow. And the trick of this growth is that you make many times more mistakes, as in the book “The Subtle Art of Pofigism”, and you learn many times faster. The stakes are higher and the result is also multiple higher.
Ivan took part in the Epic Growth Story project, a flash mob from Gett and the Epic Growth Conference, where product and marketing experts share epic growth stories and growls. A selection of the best stories: egconf.ru/growthstory and in the Telegram channel t.me/epicgrowth .
