Batyana-kombat for the developer: specially trained dudes, who are like consilers in the mafia

We have a lot of developers here in the Workshop. For the most part, many have already learned not to eat dried pizza at night, talk to real people, and generally run their business. Sick place, of course, was to obtain specialized knowledge. In the sense that it’s understandable where to code, but what about the project as a whole is not.

In Silicon Valley, the normal method is to call a specially trained guru who will give accurate and correct advice, after which everything will become clear. And you can continue to work confidently.
')
Now I will tell you how such things helped in the development and around it. In general, the main problem of the psychology of programmers who have gone into business is the focus of attention on the constant turnover and lack of understanding of priorities.

Two more commits, and we will start doing business! From Monday!

So, the big trouble of projects under the leadership of developers is that they are drowning in routine. At the beginning of the project, everything is enthusiastic, and it seems that "Here, we will work half a year now, and everything will be cool." Half a year passes, a year does not get cool, but it becomes harder because you do, do, do, and the result is not visible. And most importantly: if you do not work with the user, most likely you will not see this result.

I understand that writing a product is fun, comfortable. You sit, write code, try new tools. It would be desirable to work, instead of checking on any completely absolutely basic functions on live users. Often there is such a thought: “All week we plowed like damned. The result is not obvious, but maybe it will someday appear. ”

It does not happen that “Now people will come and find out about us, our products will sell themselves”. Usually, when it seems that the product sells itself - this means that the team has worked long and hard. And from the outside it seems that the product seems to be spreading by itself, virally, but in fact, the guys are just doing the right things.

Need debag

If the project is first, second, third, then it meets with a lot of difficulties at all stages. Therefore, the founders themselves often invite special trackers - analysts, who are doing the “debugging” of real things and looking for the narrowest place of work; or architects to help figure out where to dig further. The role of the tracker is a lot - from the answer to the question “why we are not growing for two weeks already” and ending with specificity.

Here is an example.

What does tracking look like: an example with cohorts

Project problem: the developer fastens a new feature, but does not understand how much users need it. The tracker looks at its metrics and gives the following: the revenue of the current week says nothing, because there are too many different cohorts - those who have just connected, those who are already from the very beginning with the project and different intermediate people. And the answer to the question is connected only with a part (and quite specific) of these people.

The tracker shows how to do the mechanics of assigning a Client ID depending on the source, and how to load information on the client into the API - its downloads, other important actions, and so on. Then shows how to transfer this ID to CRM. Here is a piece of dialogue:

Tracker : Client ID is assigned by Google, on the site for the first time, the first time it arrived, it received a Client ID. Client ID you ask, you don’t climb into a cookie, you ask java-script on the site: “Google, what is its Client ID?” - “Such and such” - “Okay”. Memorized in custom dimension.
Developer : Custom dimension is what? I myself take from Google Analytics, put in Google Analytics?
Tracker : Yes. What happens next? When he registers with you, he downloads, for example, you ask again: “What is the Client ID?” - “This and that”. When he registers, you write to the database: “This dude, he entered the email to you, and he had this kind of Client ID from Google Analytics, and you send an event to Google Analytics:“ Dude with this Client ID, also with such a User ID, because you got his email, registered ”. Out event in Google Analytics later.
Developer : And then you know the whole story?
Tracker : Then he buys, he buys from you at all in CRM, but when you registered him, you put yourself in the CRM Client ID when he registered. In the CRM, the purchase event happened, the event is sent to Google Analytics, you remember its Client ID: "Dude with such a Client ID, with such a User ID ...".

Further - even more integration history, which still needs to be done between different tools. Naturally, the developer himself did not understand how it works, and did not know where to dig - and the tracker just showed on the fingers how to adjust everything and what you need to have as a result. Then the developer himself figured out the API and made the desired implementation. Plus, in parallel, I used a couple of useful tips and immediately wrote down to myself competent analytics not only of the initial question, but also of other key metrics that surfaced in the dialogue.

In general, questions to trackers are of three types:

1. “See what feature we came up with. We will code by the end of the week. What to do with it next? "
2. “Something we have not grown for half a year. There is enough life, but I want business growth. We code-code, but the result is the same. What to do?"
3. "A crisis! What to do?".

Hello, senior friend

The tracker is not involved in the project, but follows him and helps with advice, connections, less often - with business. As a rule, by experience, the relationship with the tracker begins to take shape like a family. Well, he, like your surrogate dad.

When the tracker is sitting in the same building, you, like a small child, run up 10 times a day and ask simple questions from the “why the light is on” series, and in response, he talks quickly and easily, so that you understand the point. Sometimes, of course, you run up to the difficult question “where do children come from?” And then the tracker with the serious face of a scientist tells you that the world is not so simple. And, again, trying to bring the main. Such conversations are most valuable.

In addition to direct support with the correct wise advice, there are many other useful things from the tracker. For example, there was a project that planned to sell advertising, but somehow it did not go. The tracker picked up and dialed the person who became the first advertiser in a couple of days. Contacts tracker - is power, because he is with a lot of people "on a short leg."

It happens, the tracker helps with absolutely banal tips, after which everything starts to work. When the developers could not start the finished service. "Dad" advised to distribute free accounts on trial to several specific people. They began to use, got the necessary profits ... and launched word of mouth. Another service started with a monthly subscription - there, again, after technical conversations, the tracker offered to give accounts for 1 day for 50 rubles - and it worked.

"Debag" project

Digging a little deeper, the tracker is the one who helps write the business development algorithm. For example, one of our residents invented this story: you need to take a picture of your wrist to get information about your clothing size. Well, you know this song, when a man buys pants, and tries to look at the tag on his back - it looks very fun. And the girls have another problem - the fact that in one place M, S may be in another - in order to add some, they say, I am thinner in this store. So, instead of allowing digging into the development, the tracker simply sent the developers to Atrium, TsUM, GUM, Evropeysky and other shopping centers to ask how often people were wrong with the size. Perhaps after this testing, the project is waiting for a sharp turn, or even two.

There are two most frequent questions for trackers - the first is that something needs to be done, but it is not clear that. There are no ideas, there is not enough knowledge and experience, time is running out and nothing is moving. The other option is better: we do a lot of things, we try, and the result is not visible. In the first case, when it is not clear what to do, there are techniques. For each type of business, for each process in the company there are experts and best practices. There are sales techniques, there are interface design specialists, architects, and so on. Often we need knowledge from the domain - for example, developing a system for restaurants is logical to constantly communicate with people from there, and not with IT people. The question is, what experts to call, and on which right now time and money do not need to spend.

In the general case, the tracker structures the situation, highlights measurable and interacting elements in business, among which you can search for a bottleneck. Then it helps to find the current bottleneck. And then he gives the necessary techniques or experts for fixing this bottleneck. Then he controls the team’s focus on the right job.

You can’t become a tracker, you can’t become a mentor, just waking up one morning is a process tied to cross-training on each other between the tracker and the projects with which he works according to the methodology. In general, in Russia, for several years now, the nutrient medium has evolved, from which the trackers have come out. The average portrait of a tracker is a man (less often a woman, but there are in the ranks) after 30, ten years connected with IT, where he developed well and achieved something in his direction (be it development, PR, marketing, management), at least tried - and then it turned out - to create your own company, the last few years working with start-ups as an expert.

We now have three things in #tceh tracker dads leading projects

The first is Phil, the Jedi of communications. Responsible for marketing Eviterra. Prior to that, for several years he was engaged in his own project of monitoring and analyzing social networks. Looks like punk: mohawk and scary shoes. But, in general, he is kind. As long as you speak in numbers.

Misha is a programmer from school . I scored on the MEPI, I went to adminin telecom and pulled the cable into one hare. Then there was a studio: Misha Kodil, another comrade drew, together they were looking for clients. Then he began to work alone in freelancing. Accidentally hit one of the first start-up parties in Moscow. There he met a couple of good guys - and they launched GreenfieldProject. In fact, one of the Mohicans who develops the whole topic of technological entrepreneurship these years. Hakatons also organized a couple of times. The quietest of the trackers (because the thought), he is responsible for the entire program of tracks in #tceh.

Zhenya - for a long time, methodically and persistently - for 11 years - developed his IT outsourcing company. Over the years, having a technical background, several times went away from business management to the technical side: because if you have a federal client and everyone is resolving bureaucratic issues, then Zhenya is in a hurry to help. And it so happened that all the technical solutions that he formed are still in use. And all the business processes - finance, sales, work with people - he also rebuilt. And he also became very corrosive - and his favorite trick - to take the project out of the comfort zone - to say: “but you don’t have it, dude”. In general, he is angry, but terribly useful.

Users are not always honest with you.

Without a tracker at first, when there is almost no experience in the field, it is hard. Many rely on feedback from users and friends in plotting the course. This position is frankly a failure - users will not always be honest with you, because people do not know how to predict their consumer behavior. “Friends liked it, but I didn’t ask further” - this is almost a sentence.

It may seem to you that the thing you are doing is bound to fly, but before the check is just a hallucination. From here one more task of trackers - to be fair with the project. More precisely, that the project manager was honest with himself. “I know that people need this feature” - this is a bad position. "I collected pre-orders" - good.

Shredder for money and the blood of young startups

If you think that the tracker helps in all three situations, then you are cruelly mistaken. Like a real Italian Mafia consilee, the task of the tracker is not to help with your own hands, but to make the Dons look at their work from another point of view. That is, in about half the cases, the tracker helps the project to die as quickly as possible. The same Misha, for example, has his hands stained in the blood of little start-ups.
Because the tracker introduces a scientific approach to the project. Building a hypothesis, testing it, discussing how to quickly and cheaply check - or crystallizing the understanding that the project is a shredder. A shredder who seizes investments, processes them into garbage and gets them again if you are lucky (it happens sometimes).

The job is not to say, “I know where the treasures are, let's follow me,” but to teach you to think through iterations. Oddly enough, what is unit testing, development iterations, and so on, the developers are well aware. But to apply the same approach to business is difficult.

90% of the effort is not there

Again, dispelling myths, in practice, 90% of the efforts of the first project goes wrong. Often most of the development is simply detrimental to the development of the project. A frequent example is a two-week developer who cuts a part that is needed by 1% of users, whereas there are hot requests for features from a whole crowd of 50%.

But it is simple. It is often more difficult to choose, for example, between “stop working on a site and make a product” or “slow down with a product; the site is a settler, work with it”.

As a rule, there is one tracker that looks at the project as a whole and assesses the feasibility of actions, and “highly specialized” ones that can help on the specifics - somewhere you need to be told by the technical part, somewhere you need an analyst and so on. Here, see the fragment from one of the tracking meetings. Here, the developers are clearly carried away with the code and the product, when, as it was possible to simply take and sell:

Here is another example.

The project manager is constantly on support. This is wrong, and he advises the tracker what to do:

- If my traffic increases, my support will drown because the support processes 60-70% of the tickets, and 40% rise to me.
- Technical support - is it your girlfriend, who is the frontline?
- Yes, frontline. She is a frontline pumped. She has more or less English skills, plus she’s a tester from the beginning, she can install, she knows Linux, Windows, can deliver all of this, and can track down errors. And she communicates with customers. I promise support only paid customers. I will explain my logic now. I have these light versions, which, in theory, without support, but if they write to me: “I downloaded the light version and can't put it, it gives an error during installation” ...
- Download the paid version, and we will put it to you.
- So tough?
- You have technical support - this is a bottleneck that is hard to embroider, because while you find people, while you train, for the time being - long and expensive. Therefore, before doing this, it is necessary to make the efficiency of the system maximum subject to this limitation. Your money influences in this place is whether the payers received support, and whether those who will become payers received support. First of all, it affects the second. Accordingly, the task is to learn to give support to those who will become payers, and not to give everyone else, that is, passengers, to cut off passengers.
- How to do better?
- As soon as you have a ticket to go up to you, it is cut off if it is free. “Okay, man, this is a difficult ticket. We are complex tickets only paid users process. If you want a complex ticket, pay the money. ” And the second is that the support template should appear.
- He himself can make a pattern.
- You have to write such a template.

In general, most of the projects still die in the first year. It often happens that the project seems to be starting to grow, and everything seems to be fine, but in fact the developers could just like the task and the way to solve it - and they didn’t even confirm the existence of the problem that they were solving. It seems that now everything is great, they are starting to think about investments, they are pouring money into marketing. And then it turns out that there is a problem, everyone acknowledges it, but if you have to pay money for its solution, then you can be patient, not so urgent. Often this ends it.

The project sometimes has time to regroup in the air, take ready-made code and developments, go into some other niche, but this is rather an exception. Therefore, it is important to know in advance what and who needs it. And here we need the experience of those who understand something about the market (owns the insiders in the good sense of the word) and - necessarily - practical tests.

"Doctor, what is the matter with me?"

Basically, trackers are needed by those who already know what they are doing (“how to dig”), but need general help as to where to dig. Given that most projects pay for tracking from their personal money, it can be argued that this is more than necessary. It is better to spend 20 thousand rubles at the start than half a year to go nowhere.

Yes, the tracker is a kind of guarantee that you will not work for a couple of months without a result. It forces you to test every hypothesis at the moment of its appearance, which is very, very useful. And in parallel - to think about goals not in the categories “let's finish the framework”. For more adult projects, they are looking for what is stopping them from growing - and helping to expand this bottleneck.

Something like that. If you are interested in more about trackers and how to get exactly to ours without sitting in #tceh - here is the link . — . , , . , , . , , : «, ». , , (, , , — , « » — , , ).