A child project of Uetke is Python For The Lab, a website where you can find articles about Python programming and a Forum to evacuate your doubts. We had also included a newsletter sign-up form, similar to the one you find here on the right or at the bottom of the page. When deciding how to develop the websites, we went for the full experience: built almost everything by ourselves.
Static websites are fashionable these days, mainly because they are very low maintenance. You can receive a high amount of visitors in cheap servers, without the risk of compromising data or infrastructure. We built our workflow on top of Pelican, a Python-based library for generating static websites. We had to build the template ourselves, including special plugins for working with header and social images, etc.
Static websites offer little interactivity with the visitors. Newsletters, comments, feedback, etc. all have to be handled in a different server. In the beginning, we had just one small box to allow users to sign up for a newsletter. However, this box was not visible on mobile phones, the platform of choice of our users. We added a second box at the bottom of the article pages, which looks exactly the same. Actually, it was the same box.
The number of sign-ups to our newsletter didn't improve. We had some very successful articles, with thousands of visitors, but not a single subscription. It is very hard to predict user behavior, so we assumed it was just that people were not interested in our newsletter. Curious enough, with much fewer visitors, Uetke.com was receiving a constant stream of sign-ups.
We have fixed that on Python For The Lab and now our numbers are increasing steadily. Considering the time this bug was around and our conversion rates, we lost a significant amount of sign-ups to our newsletter. Our core business lays somewhere else completely, therefore it wasn't a big deterrent. However, such a small mistake can have a huge impact if it happens in the wrong place.
If that newsletter would have been a central part of our strategy, it would have been a gigantic setback. Investing in professional support would have paid off in no time. Scientific software plays the same role.
Header photo by Estée Janssens on Unsplash