March 17, 2013 at 04:43From the history of one startup
By the will of fate, launching another project, I came across a rather interesting fact.
Many of my friends try to do projects in one way or another, but few get it.
I want to tell you with small examples about what needs to be done in the project, and what can be scored, even if it contradicts your intuitive aspirations.
The following is not a recipe for all occasions, is my personal experience and does not claim to be absolute truth. You can always saw the ingenious nginx for two years, which will then take off without any marketing due to its exceptional quality and gain popularity, but the likelihood of making such an IMHO product is even lower than a successful typical startup.
So, we started to do a project of an association of companies, and special offers were to become the center section. Each company has its own website, as well as one or more successful projects, and there are promotions and discounts.
The site was supposed to combine these special offers and discount offers in the style of a group, and on the projects themselves, remove sections by inserting widgets there. Programmers know the principle of DRY / SPOT, this also works in project management.
I will say in advance - the project became successful in less than a month and paid for itself, began to earn money, but more on that below.
There were four people in the team: two managers (I, as a consultant, formerly a programmer, and the project myself, formerly a designer), designer and programmer. The designer and layout designer were involved in piecework.
The first thing I would like to note is the features of professional deformation. This is one of the first reasons for the failure of the projects. The bottom line is that in case of uncertainty, every specialist, either at the beginning of the project or at the time of a pause in development, is drawn to do what he fumbles with. The programmer sits down to write the perfect code or refactor the existing one, the designer - to redraw, and so on. While problems usually lie in a different area, and their omission leads to the failure of the project.
In this case, at the initial stage, we also began to slip a little - we thought out a complex ideal form for sending an application for a share from the site, the project designer was drawn to redraw something.
But we pulled ourselves together and did what needs to be done every time.
Firstly, we set the digital goals of the project under the designated vector. The vector was to create a stream of applications for stocks from the site, which were then transferred to sales for processing.
Digital goals were as follows: the minimum number of hosts per day, the minimum number of views per visitor, the minimum number of applications per day. For each indicator, there are three ranges of values - satisfactory, good and excellent (the ideal is taken as one hundred percent, 75% exc., 50% chorus, 25% beats). And conversion at the rate of 0.5 percent minimum - 100 hosts give 0.5 requests.
As a second step, we made statistics on the site in addition to standard tools, which allows you to count clicks on all tools and see the entire cycle - from where the client came from, what he clicked and watched, to the identifier of the requests sent to them.
This, too, was contrary to the usual notion that you just need to let traffic go and that’s it. In the future, due to measuring the number of clicks before the application, we came up with a lot of things, such as the stock of the day, selecting individual stocks, transferring the application form from a separate page directly to the stock page to reduce clicks, varying the names of the submit button, a / b testing of different text options. It is simply impossible to generate ideas (and steal too) and measure their effectiveness without fine detailed statistics.
Next, we described which sites and how we will increase traffic. Gathered statistics to analyze the effectiveness of traffic sources. In the process of startup work, we, measuring effectiveness, also tried different advertising options. It turned out that specific offers of stocks on banners were many times more effective than the general words “special offers for travel agencies”; narrowly targeted mailings to customers from the CRM database for paying customers who might be interested in certain stocks due to the structure of their past purchases were also effective.
The move went perfectly when a thematic plan was sent out before March 8, and when the last day of certain actions was announced. With the usual five applications per day, such moves allowed receiving up to 20 applications, and about 10 percent of customers bought three or five shares at once, sending several applications from the site.
Extremely unusual were the daily releases for the programmer. It was also difficult for us all that the site did not have text on many pages, it had an imperfect design and layout. This contradicted the idea of filling the site and licking the design, which everyone wants so much. However, applications were coming from the site, customers were buying, and this was pleasing.
Another interesting feature was the simplification of the interface. Consider the application form as an example. Initially, there were fields in it: the project to which the share is tied, the list of shares, name, phone, e-mail, comment. After launching the project and the first hundred hosts, no requests were sent for three days, and I decided to simplify the interface. At first, this caused protests in the team. I left: all shares are in one list, name (optional), and phone / e-mail (in one field). And the applications have gone!
After that, we wrote a plan for pumping traffic when it became clear that a hundred hosts were few. And developed up to 250 hosts per day, 800 views on average. And for the month of the project’s work, more than 120 applications have already been sold, services have been sold for several hundred thousand rubles!
Perhaps the story is somewhat messy, so I'll try to summarize:
1. At the time of the launch of the project, an important is a narrow scope, vector, the task that it must solve.
2. It is important to clearly understand the audience (in our case, it is travel agencies that can buy something at a discount, usually from the regions)
3. It is important to write down the digital goals of the project, analyzing which every day and even an hour, you will finish the project
4. Secondary elements of the site that are not related to the narrow task - design, content, beauty and little things are not important.
5. Of crucial importance is understanding the behavior of the client on the site before each click and keystroke, you need to make an effort for the most accurate tool for measuring the behavior of your visitor
6. Marketing efforts to attract traffic, and finalizing the project by analyzing statistics and the dynamics of digital goals consumes up to 80 percent of the time. Design, layout, programming in the first stage should work on this point, especially programming - changes should be made as quickly as possible.
7. After reaching the stage of proof of concept (we had the idea of “twitter-like feed of special offers from many sites through which they will order”), you can choose the second vector, giving programmers time to refactor, people to fill out the site with content, designers to save shadows and trendy metro ui redesign. And they themselves are already a new section.
We took the second concept, code-named "electronic manager."
I wish success to everyone who makes projects, and I hope that something has helped!
Many of my friends try to do projects in one way or another, but few get it.
I want to tell you with small examples about what needs to be done in the project, and what can be scored, even if it contradicts your intuitive aspirations.
The following is not a recipe for all occasions, is my personal experience and does not claim to be absolute truth. You can always saw the ingenious nginx for two years, which will then take off without any marketing due to its exceptional quality and gain popularity, but the likelihood of making such an IMHO product is even lower than a successful typical startup.
So, we started to do a project of an association of companies, and special offers were to become the center section. Each company has its own website, as well as one or more successful projects, and there are promotions and discounts.
The site was supposed to combine these special offers and discount offers in the style of a group, and on the projects themselves, remove sections by inserting widgets there. Programmers know the principle of DRY / SPOT, this also works in project management.
I will say in advance - the project became successful in less than a month and paid for itself, began to earn money, but more on that below.
There were four people in the team: two managers (I, as a consultant, formerly a programmer, and the project myself, formerly a designer), designer and programmer. The designer and layout designer were involved in piecework.
The first thing I would like to note is the features of professional deformation. This is one of the first reasons for the failure of the projects. The bottom line is that in case of uncertainty, every specialist, either at the beginning of the project or at the time of a pause in development, is drawn to do what he fumbles with. The programmer sits down to write the perfect code or refactor the existing one, the designer - to redraw, and so on. While problems usually lie in a different area, and their omission leads to the failure of the project.
In this case, at the initial stage, we also began to slip a little - we thought out a complex ideal form for sending an application for a share from the site, the project designer was drawn to redraw something.
But we pulled ourselves together and did what needs to be done every time.
Firstly, we set the digital goals of the project under the designated vector. The vector was to create a stream of applications for stocks from the site, which were then transferred to sales for processing.
Digital goals were as follows: the minimum number of hosts per day, the minimum number of views per visitor, the minimum number of applications per day. For each indicator, there are three ranges of values - satisfactory, good and excellent (the ideal is taken as one hundred percent, 75% exc., 50% chorus, 25% beats). And conversion at the rate of 0.5 percent minimum - 100 hosts give 0.5 requests.
As a second step, we made statistics on the site in addition to standard tools, which allows you to count clicks on all tools and see the entire cycle - from where the client came from, what he clicked and watched, to the identifier of the requests sent to them.
This, too, was contrary to the usual notion that you just need to let traffic go and that’s it. In the future, due to measuring the number of clicks before the application, we came up with a lot of things, such as the stock of the day, selecting individual stocks, transferring the application form from a separate page directly to the stock page to reduce clicks, varying the names of the submit button, a / b testing of different text options. It is simply impossible to generate ideas (and steal too) and measure their effectiveness without fine detailed statistics.
Next, we described which sites and how we will increase traffic. Gathered statistics to analyze the effectiveness of traffic sources. In the process of startup work, we, measuring effectiveness, also tried different advertising options. It turned out that specific offers of stocks on banners were many times more effective than the general words “special offers for travel agencies”; narrowly targeted mailings to customers from the CRM database for paying customers who might be interested in certain stocks due to the structure of their past purchases were also effective.
The move went perfectly when a thematic plan was sent out before March 8, and when the last day of certain actions was announced. With the usual five applications per day, such moves allowed receiving up to 20 applications, and about 10 percent of customers bought three or five shares at once, sending several applications from the site.
Extremely unusual were the daily releases for the programmer. It was also difficult for us all that the site did not have text on many pages, it had an imperfect design and layout. This contradicted the idea of filling the site and licking the design, which everyone wants so much. However, applications were coming from the site, customers were buying, and this was pleasing.
Another interesting feature was the simplification of the interface. Consider the application form as an example. Initially, there were fields in it: the project to which the share is tied, the list of shares, name, phone, e-mail, comment. After launching the project and the first hundred hosts, no requests were sent for three days, and I decided to simplify the interface. At first, this caused protests in the team. I left: all shares are in one list, name (optional), and phone / e-mail (in one field). And the applications have gone!
After that, we wrote a plan for pumping traffic when it became clear that a hundred hosts were few. And developed up to 250 hosts per day, 800 views on average. And for the month of the project’s work, more than 120 applications have already been sold, services have been sold for several hundred thousand rubles!
Perhaps the story is somewhat messy, so I'll try to summarize:
1. At the time of the launch of the project, an important is a narrow scope, vector, the task that it must solve.
2. It is important to clearly understand the audience (in our case, it is travel agencies that can buy something at a discount, usually from the regions)
3. It is important to write down the digital goals of the project, analyzing which every day and even an hour, you will finish the project
4. Secondary elements of the site that are not related to the narrow task - design, content, beauty and little things are not important.
5. Of crucial importance is understanding the behavior of the client on the site before each click and keystroke, you need to make an effort for the most accurate tool for measuring the behavior of your visitor
6. Marketing efforts to attract traffic, and finalizing the project by analyzing statistics and the dynamics of digital goals consumes up to 80 percent of the time. Design, layout, programming in the first stage should work on this point, especially programming - changes should be made as quickly as possible.
7. After reaching the stage of proof of concept (we had the idea of “twitter-like feed of special offers from many sites through which they will order”), you can choose the second vector, giving programmers time to refactor, people to fill out the site with content, designers to save shadows and trendy metro ui redesign. And they themselves are already a new section.
We took the second concept, code-named "electronic manager."
I wish success to everyone who makes projects, and I hope that something has helped!