Geekly Articles each Day
📜 ⬆️ ⬇️

PVS-Studio. Counting prices in the opposite direction

PVS-Studio cost

If you go to the purchase page of PVS-Studio, you will see that we have removed the price and offer to contact us to discuss it individually. So far this is an experiment, the result of which we cannot predict. The disappearance of the price does not mean its growth. On the contrary, we want to offer PVS-Studio, including at a lower price, for small teams. If you are interested in how prices have changed and why we want to try out such a sales model, then I wish you a pleasant reading. I think those who are engaged not only in development, but also in sales of their products, will find in the article something interesting for themselves.


')

How it all began


Initially, we (Andrey Karpov Andrey2008 and Evgeniy Ryzhkov evgeniyryzhkov ) wanted to create one or several small products, in the spirit of shareware. A soul full of motivational books demanded action. Thoughts were different, but for a long time nothing was invented. I did not want to make another clone of something. If at that time iPad and other tablets were already distributed, I think our choice would be obvious. It is not known how an alternative story would turn out, where we would try to make games or something else. And now we will not know. But at that time it was not easy to invent something. The iXxx industry was not there yet. Sites do not want to.

And we came up, as it seemed, a great idea. We decided to make a simple code analyzer that would help programmers to port their applications to 64-bit versions of Windows. Relevance was fully justified. Prior to this, in our previous work, we faced a situation of extreme difficult catching fleas in programs that began to use more than 4 gigabytes of memory. Tools simply did not exist. At all.

It seemed to us that such a product is doomed to success. At that time, there were rumors that after one, maximum two generations, Windows would become exclusively 64-bit. This will entail a sharp wave of demand, and we will be able to make a good profit during this period of mass migration. And then, we would think about doing other projects.

But man proposes, and Microsoft has the god. The transition from Win32 to Win64 stretched for a huge period of time and until the end of Win32 support is visible. As a result, no surge in demand for such programs has arisen. Interest was and is, but very weak. So, we can say that the first product of Viva64 for searching 64-bit errors did not bring noticeable profit. Rather, it turned out to be 0. And at first, I generally had to work as an outsourcing.

The following two directions were even more unfortunate. It was an analysis of parallel programs built on the basis of OpenMP technology and the code parsing library VivaCore. In 2008-2009, there was a lot of talk about multi-core architecture. We are by nature trusting people. We believed that a little more, and without parallelism will not live. And the queen of parallelism will be OpenMP. This also did not happen.

Based on the VivaCore library, we planned to make specialized tools to order, but did not receive any interesting offers. And we slowly turned this direction. No, there were some communication. But they were in the spirit of: “Make us such a specialized code analyzer and give it to us along with the source code. And we will pay you 50,000 rubles. And maybe even as much as 100,000 rubles. ” It was not even funny.

The first successes were waiting for us only with the advent of the PVS-Studio tool. At the beginning, it included the analysis of 64-bit programs (Viva64) and OpenMP programs (VivaMP). To popularize and out of despair, we began to do a general-purpose analysis slowly. At the beginning, this part of the analyzer was intended to be free. It had to exist to attract programmers to paid functionality (Viva64 / VivaMP). We did not believe that in the coming years we could at least somehow compete in a general-purpose analysis with Gimpel PC-Lint, Coverity, and other sharks of various sizes.

And then, we wrote a man. He asked whether it was necessary to buy PVS-Studio in order to use general-purpose analysis. We replied: “No. It is free. But we will be happy to sell him Viva64 / VivaMP and provide support. ” Man: “Oh no, thank you. If it is free, I don’t need anything more. Thanks for the nice free product. If I had to buy a general-purpose analyzer, I would have bought it. I like him. I have no need for Viva64 / VivaMP ”.

And here in our head clicked. Here it is. It is interesting to people. And the first release of PVS-Studio 4.00 is already released with a paid general-purpose analyzer. Since then, it is general-purpose diagnostics that lead us to the most buyers. At the same time, it was much easier to promote them. The examples of errors found are simple and understandable.

All this, of course, is of little concern, but I think it was interesting. And now let's talk about prices.

Price history


As I said, we initially thought of a simple mass-consumption tool. The initial price for Viva64 was about $ 400. But then began a long revision of the price:
It really is a degree of two. The reason for these changes is that the product was not massive. Moreover, we were constantly convinced that a large company is not so important value. And the main interest from the very beginning was just from the side of big companies, not small teams or individual developers. For the same reason, we abandoned the single-user license.

It happened like this. Large companies come to us and buy a license for one (ONE!) Developer. They motivated such a purchase for various reasons. For example, it could look like this:

The duties of tracking are entrusted to one person. He reviews the results of the analysis once a day and, if he noticed something, sends messages to the bug tracker by the appropriate programmer. And it does not matter that the team is big. We, please, for one person.

It is not clear how to challenge this position. After the next such incident, we simply removed the single licenses. Yes, we lost small teams, but still it was more profitable than convincing big serious teams not to be greedy. From the side it is difficult, probably, to feel. I will try to give at least some analogy.

Imagine you are connecting gas to people at home. You take it cheap, because not so much gas flows through the pipe to the house. Cakes are baked, turkey is fried. For God's sake. And then, someone in the building instead of gas stoves for cooking, opens a thermal power station. And he no longer plans to pay, arguing that he also needs only one pipe. Of course, it would be appropriate to talk about the volume of gas. It is a pity that the counter to check the code is a difficult philosophical task. About this I will talk a little further. In the meantime, I think you understand why, after the next such connection of TPPs, we abandoned the single pipes (licenses). It turns out more profitable to sell a little, but expensive. Sadly

The last prices we had on the site are:
Naturally, the main purpose of such licenses and prices is to sell the Site License to large companies. And we sell such licenses. Little, but sell. However, the situation itself does not please us. It turns out that we have lost all the small and medium teams. For them, such a purchase is unjustified. Of course, making the version now again for $ 500 is not an option. I will not give here all the arguments and calculations. Take a word - we will earn less and support more people. However, leaving everything as it is is bad. This is where our searches come from.

Ideal approach to pricing


Assigning differentiated prices for projects of various sizes seems to be a simple and correct solution. I gave the analogy with gas above. The more a person consumes gas, the more he pays. The more code is analyzed, the larger and more serious the project. The larger the project, the more money it is reasonable to take from the team.

That's just the trouble to come up with such a counter. It is not clear what and how to count. Of course, in the first approximation, it seems that it is easy to calculate certain units of the size of the project, and to set the price on this basis. During the year we thought a lot and discussed this area. However, the more they thought, the more complicated everything turned out. There are an incredible amount of nuances that turn a simple and understandable idea into something unintelligible and difficult to implement. All this will not be easy to explain to a potential user. I will not dive deep into this topic. I will cite only a few points, pondering each of which, you can feel the complexity of the problem.


I am sure that many will want to share their ideas on the topic of calculating the size of the project. Please do not do this. We have been thinking about all this for a long time and could not come up with an adequate and convenient solution. If I do not respond to the ideas put forward, it may seem rude. I do not want to enter into long discussions. We talked about it anyway. So just believe. Yes, this calculation can be done. But to make it simple, elegant and beautiful is not.

In this case, such a calculation will still rely on the credibility of users. So it's easier not to fence anything, but, having communicated with people, to understand the size of their team and the tasks assigned to them. And who wants to cheat, still deceive.

It is logical to make different prices for different size projects and different size teams. However, we did not make a large table with prices for all occasions. We want to try to first communicate with a potential client and offer him the best option. Perhaps in the future we will again change the format of the order page. However, now we want to try just such a way to sell and negotiate prices.

What is now with the prices of PVS-Studio


Having read all of the above, the reader may ask a reasonable question: “So what are the prices for the PVS-Studio code analyzer now?”. I answer.
  1. The price for full-sized companies with a large project and a team of up to 9 people, the price will be around € 5,250. Around - because someone may want to implement certain wishes. Please note that we are talking about the size of the team, and not the number of users of the program.
  2. Larger development teams or those who need extended support should purchase a Site License, which starts at 9,000 euros.
  3. Small teams can count on a discount. The discount depends on many parameters. Write to us - let's talk.
  4. We support free open-source projects and give out free keys for a while. To give out the keys for a longer period is a burden for us, since all the same, support is required over time. After all, these users will become more and more. Therefore, for them we can offer very inexpensive options. In general, write - we will find a compromise.


I want to emphasize: our goal is to build on the size of the project / team, and not on the number of installations. Therefore, the size of the team refers to the size of the team, and not the number of people who will directly launch PVS-Studio.

We are waiting for everyone who is interested in mail. We are ready to discuss various purchase options.

The trial model remains the same. The version has full functionality, but is limited by the number of "clicks". By the way, PVS-Studio 5.10 will be released soon, where we will present the first version of the Standalone version. It will be able to check * .i files generated in any way. This should significantly expand the range of possible applications. Also, Standalone version contains an interface for working with a list of messages and the text of the program. All this is only for Windows. However, already a big step forward.


Sincerely, your dear unicorn.

Source: https://habr.com/ru/post/196312/

More articles:


All Articles