In September 2018, after a few weeks of developing and user testing, the very first version of Flick was launched, and with it, the ‘graph view.’
When we introduced Flick, it was a part-time, experimental project that we hoped to utilize to support our agency business. At the time, the priority for myself and my co-founders was to launch a version of our product as quickly as possible. We stripped away everything that was not necessary at that point in time – So much so, that we did not even have a name for our product and it was initially marketed as “The Hashtag Tool”.
The first version of Flick needed to allow users to find related hashtags to their search term and display a handful of key metrics to help them pick hashtags that would be compatible with their account. We built a working prototype using Neo4J’s bubble graph.
This allowed us to collect feedback within a week of starting to code, which meant we could begin validating our ideas and developing a product roadmap. At this point, Flick had no real user interface and needed to be run using Neo4J.
Naturally, the next step was to build a front-end that allowed users who did not know how to code, to use the platform. With speed being our top priority my co-founders and I opted to borrow the design from Node JS and output search results as a graph showing the relationship between hashtags via different bubbles:
Once this was done, we had a product ready to launch and iterate on. What we did not anticipate was the popularity that The Graph View had with our early adopters. What was originally meant as a shortcut, quickly became one of Flick’s early selling points.
With that said, over the last two years, Flick, and our users have evolved massively. The rest of this blog post aims to provide some insight into why we have decided to retire our beloved Graph View from Flick.
Why we are sunsetting The Graph View
Throughout the development of Flick, the team has looked to continuously improve the product, with changes being made on an almost daily basis. Over the last 12 months, we introduced two new ways of viewing search results:
The Table View was designed for users who heavily rely on indicative metrics (e.g. Average Likes, Competition Score, etc.) to make decisions on which hashtags to use. It allows them to find and compare hashtags extremely quickly.
The Group View was built for users with slightly less experience, looking for a simplified tool to carry out hashtag research. It strongly facilitates the implementation of the staircase strategy, whilst abstracting all the detailed metrics away.
Reduced usage over time
At the time of launching the Table and Group views, we hoped that the different formats would complement each other. This was the case for a while, where we saw different user types use the view that better suited their workflow.
However, as Flick grew in popularity and was mentioned in more blog posts and YouTube tutorials on Instagram strategy, we began to attract two types of users. On one hand, many new Instagram users joined Flick in the hopes of having a tool that would help them understand hashtag research. On the other, very experienced users were joining Flick to streamline their process.
The newer users found the most value in The Group View, where all complexity is abstracted away. Whereas the experienced users love The Table View for all the insights they need to make key decisions on which hashtags to use quickly.
This left The Graph View in an awkward position where it just wasn’t fulfilling the needs of any user type, leading to a gradual but steady decrease in its popularity.
Today less 2% of searches made on Flick are made with The Graph View. When looking into this number in more detail, only 5% of new users ever use The Graph View.
Our team is always looking to streamline our users’ experience with Flick. We believe that if something is not bringing value to our users then it should be removed.
Poorly written & inefficient code
On another note, there are added benefits of removing The Graph View from Flick’s codebase. As mentioned above, the priority at the time of initially launching Flick was to build a working version as quickly as possible. Corners were therefore cut. Neither myself nor my co-founders ever thought we would have over 10,000 active users on Flick.
The high volume of users, combined with the rushed code has led to bugs appearing in other parts of Flick. These bugs could largely be fixed by removing The Graph View from Flick. Removing a part of Flick used by 2% of our users to improve the experience for 100% of our customers is an opportunity too good to pass up.
Ultimately, continuing to develop and build The Graph View is not something the majority of our userbase would benefit from. This was a difficult decision given the emotional attachment the team has with the very first set of features that were built on our hashtag tool. However, we welcome this change with open arms and are excited to keep building amazing features for you to use and love. I’m extremely grateful to every Flick user that has helped us directly or indirectly improve our product over the last two years. I look forward to serving you for many years to come.