This article was originally published (in Russian) at the website vc.ru. The article was translated and published at our blog by the editors' permission.
Today, content marketing is an integral part of company promotion over the Internet. A company should regularly inform the world about itself, share the vision of the company and provide useful information for the target audience. The latter will create additional value of the product for the people. In this article I'm going to talk about my experience of promoting PVS-Studio using content marketing and a side project.
PVS-Studio team has been successfully using content marketing to promote PVS-Studio static code analyzer for several years. We realized that this type of marketing is a must for us because of a very specific target audience that isn't responsive to usual ways of marketing such as banners, paid search and newsletters. Our audience is development companies, working on C/C++. Hopefully, soon the range of our customers will expand by the development departments, coding on C#, as the work on C# analyzer is in full swing.
We regularly analyze open source projects and find bugs with the help of PVS-Studio analyzer. The number of bugs we found really proves the necessity of using a static code analyzer. We also write articles about the analysis results in our blog and post them on web-sites with high concentration of C/C++ developers - such as habrahabr, reddit, isocpp, intel, hacker news, gamedev, gamasutra. This method brings people of our target audience who download, test and what's most rewarding, buy our product.
So it is an approved way of promotion for us, but as it is said "Every soldier carries a marshal's baton in his pack". That's why I decided to try something new.
I came across a very inspirational and thought-provoking article on marketing via side projects. "Crew" company, whose main business is a resource that you can use to search and hire freelance teams. These guys created an additional resource to publish free high - resolution photos (unsplash.com) and placed a link to their homepage - crew.co. This brought huge traffic to their main web-site from a side project unsplash.com. It made me think that probably we could do something like this for C/C++ programmers and of course, implicitly promote PVS-Studio :)
The main thing that was pushing people off from reading our articles was the commercial style and a considerable amount of time you have to spend reading them. But at the same time, we noticed that most PVS-Studio clients had read at least one article before they bought or just downloaded the product. So we had to come up with something valuable for the programmer audience, but also not to forget our ultimate goal - bring target traffic to viva64.com. That's how CppHints.com came into being.
The main requirement that CppHints.com had to meet is that the project should be of interest to any C/C ++ programmer. This condition would guarantee that CppHints brings only target audience to the main web-site. Within the framework of the project we decided to share tips on best practices of programming in C/C++ and thus attract an audience of C/C++ programmers. As we had already checked more than 200 open source projects, we gathered quite a number of good and bad programming practices, so we just needed to put it all together. The choice to make this project in English was determined by the fact that 80% of our customers live in the USA.
We also decided to publish one tip every day, making it concise, easy-to-read and to add a "subscribe" button to get these tips via e-mail. This was the main idea of our new resource-CppHints.com. Our idea was that these tips would rouse interest of a wide audience of C/C++ programmers and attract more than 5000 subscribers to our resource.
Everything was set up very fast with the help of outsource. We just needed a landing page and a simple administration panel - all lean and mean. The first version of the page:
Initially we planned that at some point we'd get tips from our readers, which would reduce our workload. In practice, we received such a tip only once and it wasn't of the right format to us. That's why we decided not to post it. We would rather write these tips ourselves to guarantee the quality of the resource, so we removed the button "submit a hint".
Nothing works better than word of mouth :) John Carmack, co-founder of Id Software, a company known for producing such games as Doom, Quack, retweeted the news about CppHints.com release:
Thanks to this retweet we got more than 1000 subscribers in one day; posting news on our usual blogs brought us another 1000 subscribers in 2 months. For example, just getting to the first page of news.ycombinator.com brought us 357 subscribers:
But by the third month of its existence the growth of the subscribers number started slowing down.
It became clear that we won't get such a viral effect as the Crew team after launching Unsplash. We have also switched to issuing one tip a week to reduce the work load of the team, namely of Andrey Karpov, who writes most of the tips. Additionally, I started paid search marketing aiming to get a constant inflow of new subscribers.
As I said, within the framework of this campaign we had a goal to get 5000 subscribers via CppHints who would regularly read our newsletters. Unfortunately, this goal has not been achieved.
But despite that we've managed to get significant traffic to our main web-site. Let's have a look at Google Analytics statistics on the referral traffic:
Sorting is done by the "goal" results, for viva64.com it is the number of the product downloads. The table shows that CppHints.com brought 852 visitors to the main site viva64.com, 14 of them downloaded the product. Only habrahabr.ru and stackoverflow.com have bigger figures.
The results are quite decent, but our goal wasn't achieved. We didn't manage to gain interest of an audience wide enough among C/C++ programmers. We expected to have at least 5000 subscribers in 2 months, as this was the criterion set to measure the product success.
Besides that there is one major problem in CppHints implementation - necessity to maintain it. PVS-Studio already had a blog and we were adding one more resource that needed to be filled with content which required more time and energy. As a result we had content marketing via a side project. But given that our technical director Andrey Karpov writes these tips himself, it wasn't cheap to maintain CppHints.com.
So you may ask a reasonable question - what prevented you from closing down this project altogether if the goals you set weren't reached? First and foremost, it was positive feedback and encouragement received from some of our readers. We regularly receive letters from our readers who thank us for the tips we issue. Of course, it inspires us to move on. We decided to give the project a second chance and, hopefully, in the long run, it'll bring more subscribers and new customers of PVS-Studio. Paid search marketing is still on, which also brings new people.
It really is, taking into account that the most successful projects are those that do not require additional investment. An ideal variant is when users themselves fill up the site with content. Then the only thing you have to do is to moderate it. This worked perfectly for Unsplash.com, as the photographers (intending to promote their work) submit high-resolution photos themselves. But at the same time we may raise a question - how targeted is the audience that comes to the crew.co site?
Our CppHints.com project didn't become so viral and it turned out to be quite energy consuming, as we have to put considerable efforts of our team to maintain it. That's why, before starting a side project I would advise to ask a question:
Will it be possible at some point to minimize the efforts to keep this side project on?
For example, moderating a project, not creating content for it.
If the answer is no, be ready for significant expenses that this project will cause. You should also think in advance of the criteria for the success/failure measurement of your side project. In our case it's impossible to measure the product success by the sales figures of it as the product realization may take several months. That was the reason why we measure the success of advertising only by the number of subscribers and product downloads.
Thank you for your attention, your comments and critical ideas are welcome!