Exploring ways to attract software to the site

annotation

This article will be interesting to those who promote their own software products on the Internet. The article is extremely practical. It is a report on some actions that we have carried out recently to promote the product PVS-Studio. I will tell you what works, what does not work and share related thoughts.

The scope of our activities

First, briefly about what we work with. We develop and promote the PVS-Studio software product. This is a tool for programmers, priced at 3,500 euros for the command version. The price is not small, but similar, although larger solutions from this area generally operate on five-digit numbers. So we are a kind of unpretentious option, for those who have limited budget.

The price is high or not, it is still clear that any shops, catalogs or sites like BitsDuJour.com are not suitable for us. Such tools are not acquired spontaneously. And in general, programmers do not roam the directories, looking for what would be so cool for them to buy from the toolkit for developers. Tools are acquired when there are certain tasks and they need to be solved.

Our promotion on the Internet means that when programmers have a corresponding task to improve the quality of the code, they will remember PVS-Studio. To do this, they need to know that our tool exists and read at least one of our articles.
')
I clearly understand that sales and the number of visitors to the site are different things. However, the more targeted visitors there are on our website, the more people will learn about PVS-Studio and more people will remember when a code analyzer is required. We work for the future. It often happens that a year passes from interest to purchase.

Formulation of the problem

Our task is to tell PVS-Studio to as many programmers as possible. We partially solve this problem by publishing articles at various sites (Intel.com, CodeProject.com, CodeGuru.com, go4expert.com, habrahabr.ru, vingrad.ru, and so on). Unfortunately, this path is limited. It is impossible to write good stuff in bulk. The number of good sites is not large and it does not work with all of them. You can of course use the services of newsletters of articles on catalogs. But in our case it is meaningless for three reasons:

1. No one reads these articles. Programmers go to read articles on familiar sites.
2. The reference mass is not interesting to us either. It seems to us that one link from sites such as Intel.com or CodeProject.com replaces the entire junk mailing list.
3. We do not make money on advertising, and it makes no sense for us to put random people on the site. And as a similar source of visitors there are big doubts.

Naturally, there is a desire to spend money / resources on alternative ways to attract programmers to our site. Of course, you can attract to articles located on other sites. But the effectiveness of such actions is difficult to measure. Therefore, we began to try various techniques that attract people to our site. Consider these methods and our conclusions regarding them.

Banner advertising on the site CodeProject

Site CodeProject does not like us and repeatedly deleted our articles. Even those who became the "article of the month." The reason is clear. They do not want us to mention our PVS-Studio product in the articles for free. And we see no point in writing articles where nothing will be said about PVS-Studio. I will not describe the details of the confrontation here. But the point is that CodeProject constantly “pushed” us towards paid services. In the beginning we had no money for it. Not at all. Therefore, we did not even consider paid options. Then some money appeared, and we decided to place a banner ad.

The result is extremely upset us. I will not paint tariffs, reports, CTR, calculations here. Immediately bring the cost of attracting one person to the site using a banner on CodeProject.

The price of one click on a CodeProject banner is $ 8.4.

At the same time, one should understand that the one who has come in is not necessarily so interested in PVS-Studio. Perhaps he just missed the mouse. Perhaps he is a php programmer in general, but he liked the picture, and he decided to look at the site.

By the way, now you will understand why CodeProject so disliked our articles and "chases us." The fact is that our articles are good. They are read with pleasure by people, and, probably, quite often pass to our site.

For example, from an unpretentious article " About size_t and ptrdiff_t " for all time 600 people passed to our site. And this is despite the fact that in the article they removed all links to us. Links are only present in the "About the Author" section below.

It turns out that the publication of this article saved us 8.4 * 600 = $ 5040.

Moreover, the people who came, much more targeted! Once they are transferred, they are most likely interested in the article, and they even read about the author.

In total, from our articles on CodeProject we were visited by 13,600 visitors. If we wanted to attract such a number of visitors with banners, it would cost us about $ 100,000. Of course, writing and publishing a good article costs money and time. However, all the same, the price of writing and publishing is much lower than placing a banner.

Conclusion

Site CodeProject is cool and great. But the proposed price of banner advertising is too high. If you are not Intel, you can also get the number of target visitors in a much cheaper way - by writing and publishing articles.

Apparently, we will never return to buying banners from CodeProject if he does not revise prices and policies.

Google adwords

Contextual advertising is good and useful, but, unfortunately, is not applicable in our case. Contextual advertising does not create demand. It only allows you to be first among many other offers. It makes sense to use contextual advertising when you sell bricks. Bricks sell many and need to break through among them. At the same time, not only by the word “buy a brick”, but also by related requests.

In our case, we do not compete with anyone on the word PVS-Studio. And displaying advertisements using combinations of the “find an error in a program” type is simply unrealistic. Such combinations are unimaginable number.

Of course, there is competition in our region. For example, it is reasonable to move forward on the phrase “static code analysis”. And there are such advertisements, and we made our own. But the trouble is that no one clicks on them.

The point here I think is that we work with a specific sphere and specific people.

Programmers, perhaps, are the most closed to noise in the process of searching for information on the Internet. I can judge by myself. When I search for how the Foo () function works and why it returns a specific error code to me, I cannot be lured using AdWords, a banner or anything else. I am closed and focused only on the search. I know for sure that I wouldn’t click on any ad, I won’t see an answer about the work of some code or function. I just need an article or forum.

Pay attention to the fundamental difference. If the query includes the word "brick", then a person with a probability of more than 50% wants to buy a brick, or at least read what to choose in the future. This person can be carried away by advertising like "high-quality brick". If the query includes "Visual C ++", then with a probability of more than 99%, people do not plan to buy this very "Visual C ++" or something else. He is looking for a solution to his specific problem and only. He doesn’t even need any tools for Visual C ++ at the moment.

We have already tried several times to work with Google AdWords and the result every time we were very upset. Invariably, we get a very low CTR. If you fix the price, then almost no one comes to our site. If you let go of the price, it rises to $ 5 - $ 10 per click.

We consider such a cost per click extremely high.

Conclusion

For our case, AdWords doesn't work well.

Facebook

The era of social networks has begun and naturally I want to try advertising in them. We tried Facebook and VKontakte.

About VKontakte will not tell. There could be quite advertised if there were more foreigners. But more than half of foreigners there are fake. If you choose a US country, then the list of people will be solid names in Russian: "Chuck Norris", "Maxim" and "Sweet Bar". We do not need such foreigners.

We tried to advertise on Facebook. Of the paid promotion options, Facebook seemed to us the most interesting. First, you can quite flexibly select the target audience. Facebook knows a lot about people, and this allows, for example, to show ads only to those who are in certain programmer groups. Secondly, the prices also seemed to us very acceptable.

The price of one click on a Facebook mini banner is $ 0.7.

Of course, again there is no guarantee that the person did not miss the mouse or simply admired the picture. However, if you compare $ 0.7 (Facebook) with $ 8.4 (Codeproject), then the difference is significant.

With people on Facebook you can try to work independently. You can communicate in groups, posting news and so on. We tried it, although not very active. Unfortunately, the result is almost zero. So paid advertising, perhaps, fully justifies itself.

We have not yet decided for ourselves whether we liked Facebook ads or not. The price is not very high, but there is no confidence in the quality of the perfect transitions. People come to Facebook to have fun. And this is hardly a good time to turn a person’s interest back to work (programming). Advertising toy pink bears here will be much more appropriate.

findings

Advertising on Facebook is clearly promising, but we are not sure about its benefits for our case.

Stumbleupon

If you are not familiar with the StumbleUpon site, then I recommend to look at it. At least there is interesting to wander. In a nutshell, this is the largest service of bookmarks, which is randomly tracked. But this takes into account your interests, and you can vote for the pages shown.

In StumbleUpon, you can promote your site in two ways.

Free option. You add links to the pages of your site. It is important that the pages should be interesting. Otherwise, the site simply zaminusyut or even banned. It was easy for us to work with StumbleUpon. We have a lot of interesting articles and there was something to add to bookmarks.

StumbleUpon website is able to generate hundreds of thousands of clicks on your bookmarks. But this only works in the case of photos of cats and articles about the death of "Bin Laden." Our achievements are much more modest. Somewhere in a year and a half, as we learned about StumbleUpon, from there about 11,000 visitors came to us. It is not enough. But if you consider that the efforts to add bookmarks are very small, then this is a very good source.

Paid option. You can buy shows of certain pages. Types of prices per impression: $ 0.05, $ 0.10, $ 0.25. From the price depends on the priority of the show. For example, if you are going to spend a lot of money, but put $ 0.05 per show, then these shows can take a very long time. And then you will not be able to provide the desired number of hits per day. To avoid this and there are more expensive rates.

We are greedy and not in a hurry. Therefore, we are quite satisfied with the option of $ 0.05 per impression. A thousand hits took about three days. Very good numbers. For a month, you can arrange about 10,000 impressions of a page, spending $ 500.

Of course, it is not clear who shows these pages. As I understand it, StumbleUpon will lead there those who have programming in their interests. However, programming is different and StumbleUpon does not understand shades (programming languages). For example, I walked around StumbleUpon for a long time. A lot of minus and plus. But still, even though StumbleUpon shows me programming articles, in 95% they are not interesting to me. I'm interested in C / C ++, and he gives me PHP, JAVA, Ruby, HTML, CSS, Perl, and God knows what else.

In the general expediency of paid use of StumbleUpon we are not sure yet. First, the free option works more or less. Secondly, the interest of the people involved is not clear.

Conclusion

The free option is worth using anyway. Just do not spam there all the pages of your site. Anyway, there will be no sense.

The paid option should be used if you are promoting something massive, designed for a wide audience.

Twitter

I will say right away that we have not tried any paid promotion options on Twitter. The official advertising system is in its infancy. There are all kinds of spam services on Twitter, but we didn’t even look in their direction. In general, we work with Twitter independently, posting interesting links to others and our articles. The people are slowly, but going. By the way, if you are interested in programming, we invite you to join: @Code_Analysis .

We believe Twitter is a very promising direction in terms of advertising. However, it requires a lot of effort. If you are not a TV star, then gathering and keeping an audience is not an easy task. You need to constantly publish something new, be ready to respond to comments and questions. On the one hand, this is not difficult, but on the other hand, we need to do this regularly. Efforts are spent precisely on regularly allocating time for Twitter and selecting interesting material. We have to be distracted from other activities and this constantly switching between tasks is tiring.

We decided to find someone who would regularly deal exclusively with tasks related to Twitter and other social networks. But so far the searches have not been crowned with success, and only I myself work on Twitter. When we find the person we need, it will allow us to make the publication of materials more regular and therefore attract more followers.

In addition to your own Twitter posts, you can use contacts with good people to attract people to the site. Creating such contacts is, of course, a difficult task, where specific recommendations cannot be given. However, one should not underestimate this direction. I will give an example.

After the publication of the article " 100 bugs in Open Source C / C ++ projects ", we were surprised and upset to receive absolutely no feedback. But it happens. Unsuccessful publication time, other unknown reasons. Then I wrote a small letter to John Carmack asking him to write about this article on my Twitter. And he did it . Thanks again to him for that. This link alone brought about 2,000 visitors to our site.

Conclusion

Twitter is a very promising means of disseminating information. They can and should be used, provided that you have time to work with him.

Reddit.com

Reddit is a social news site. You can post a link to something and people will appreciate it. At the same time there are many subsections, which is very good. News about programming does not intersect with the news about the release of iPad3. This is a significant advantage over Digg, for example. There it is impossible to break through with news that does not cover a large group of people. And on Reddit, you can find and bring information to the few who are interested in C ++ programming. In fact, there are a lot of programmers there, but compared to the group where the pictures are discussed, their number is insignificant.

Note that Reddit is a very specific resource. It is impossible to guess how this or that news will behave. Depending on the time of day, flares in the sun, holidays, currently sitting on the site of moderators and users, the result may be completely different. The news can be banned, sluggish to discuss, or vice versa, everyone can like it. At the same time, very little depends on the quality of the material. An interesting article may be completely ignored, but nothing unremarkable article, can give rise to a storm of discussions. In general, the successes and failures of Reddit should be taken calmly, as the whims of nature.

I will give an example to show the power of chance. For example, here on this obscure article on the day of publication came 6,000 people. But these wonderful lessons for all the time after the publication came only 4,500 visitors.

Reddit website allows you to post paid links. We wanted to try this type of advertising, but it turned out to be impossible for technical reasons. As stated here , Reddit accepts "only US, UK, and Canadian credit cards" for payment. Despite the fact that the “Redditors are from everywhere!” Response administration: “We know. We don't like it either. ”

Conclusion

I definitely recommend the Reddit site to everyone. It can generate a huge number of visits to your site. But it should be understood that people need to offer something really interesting.

Hacker news

It's about http://news.ycombinator.com/ . When I first saw this collection of news, I decided something just another pointless cesspool of links. But the first impression was deceptive. Despite the spartan interface of the site and the huge number of links placed there, this site can be very useful. Many news from there get on Twitter of various people, on the same Reddit and so on. However, the probability that your link will be ignored is also high. The speed of adding news is huge, and if the link does not have time to interest at least a few people within an hour, then no one will see it anymore.

From my own experience I can say that most of the published links go unnoticed on Hacker News. However, if you are promoting something technical, then Hacker News can be very useful.

Conclusion

You can use to promote something on IT-topics, but you should not expect miracles.

Conclusion

Summarizing the information about how much it will cost to bring 1000 visitors to the site, if you do not waste your time and using only money:

• 1000 visitors from the CodeProject banner: about $ 8,000.
• 1000 visitors with AdWords: $ 5,000 - $ 10,000.
• 1000 visitors from a banner on Facebook: about $ 700.
• 1000 visitors with StumbleUpon: $ 50 - $ 250.

Naturally, we must understand that these are different visitors. Also it is necessary to take into account that the prices quoted relate exclusively to our situation. I remind you that we are promoting a highly specialized product for programmers. Your picture may be completely different!