I run Feature Upvote, a B2B SaaS product that offers feature request tracking via a simple online board, which can be made public or kept private.
Each January I reflect on how our product and company has evolved in the last year. Here’s a summary of how we did in 2022.
(If you are a business owner, I highly recommend writing a similar “year in review” article and making it public.)
We added a marketing person to our team
For the first years of Feature Upvote, I led marketing myself. That kind of worked, but my background is in software development, not in marketing.
Truth is, I get frustrated with marketing. Unlike coding, the results are not immediate and it is hard to attribute results to any specific marketing effort. And when I get frustrated I find it hard to stay motivated and keep at it.
I was neglecting marketing, always favouring product development instead.
And so, in January 2022, we hired our first marketing person.
Handing over my marketing responsibilities has led to better and more consistent work in this area. It has also freed me up to put more attention into improving our product, where my real passion lies.
We migrated our website to WordPress
Until March 2022, our website was built using Jekyll, managed in GitHub, and hosted on Netlify. This is a classic coder’s approach to website hosting – it appeals to the coder’s sense of aesthetic because it gives us precise control over the entire HTML and CSS that the site is built on.
But this approach becomes cumbersome when you want to hand over content management to a non-developer.
So we converted the site to WordPress, easily the world’s most popular website CMS (content management system). I don’t enjoy using WordPress, but that doesn’t really matter. Our marketing person – and anyone who will help in the future with our content – can now create and edit our website content directly, without having to master developer-oriented tools like Jekyll.
I’ve started to think that we should have used something like WordPress right from the beginning. It would have saved plenty of developer time that was spent converting other people’s content and edits into Markdown.
Support is now everyone’s job
I have always personally handled our customer support. This bothers me; it keeps me close to our customers, but it doesn’t seem sensible for support to be solely done by the founder, especially now our product is a few years old.
Because I do many things other than support, I’ll admit that my efforts at support have been acceptable – but not awesome – and somewhat inconsistent.
I have long wanted to hire a support person. But we get few support requests, typically an email or two per day, so hiring a support person doesn’t make sense.
In recent months we’ve taken a new approach to support. Now everyone on the team helps a little with support. I was pleasantly surprised to see that our team was excited with this change.
We’ve also lifted our ambition with support. During our regular business hours, we aim to respond meaningfully to support requests within one hour.
Our aim is not just to answer support requests quickly, but to impress you, our customer, with our friendliness, our helpfulness, and how rapidly we solve problems.
I’ve been very happy with this change.
We reluctantly replaced our web framework
The Open Source web framework used by Feature Upvote ever since we started in 2017 had become abandoned. We were doing our best to work around this, but we found ourselves rapidly approaching a dead end.
That left us at risk of becoming susceptible to new security exploits and compatibility-breaking changes in modern browsers. We needed to replace the web framework that Feature Upvote is built on.
I’m firmly in the “almost never rewrite your code” camp, but this time it was unavoidable.
We spent a couple of months weighing up options and doing small proof-of-concepts with alternative web frameworks. Then we spent another couple of months doing the hard work.
The trouble with this type of work is that while it is happening, you can’t make any progress with new features. Furthermore, it is inevitable that new bugs are introduced during the conversion work. So you temporarily have a product that is worse than before.
This was painful but needed to be done.
We eventually settled on Javalin, because:
- it has a similar approach to the old discontinued web framework we were using
- it is actively maintained and developed
To help ensure Javalin stays active, we are financially sponsoring it.
We made our newsletter much better
Companies like ours, with B2B SaaS products, typically have dull email newsletters. They’re a simple list of new features, or just promoting new blog posts. They are all about the needs of the company sending it, and not about you, the recipient.
I wanted our newsletter to be better than this.
My aim is that, amongst the dozens of product email newsletters you receive each month, ours will stand out as the one you always look forward to reading. Something that is interesting, well-written, and consistent.
But this is hard to achieve.
We assessed what others were doing with their email newsletters. We looked for shining examples of newsletters for B2B SaaS products done well, and learnt from them.
We sent an email newsletter each month in 2022. We would critically assess each email newsletter before it was sent, and again after it was sent. In my opinion, each month our newsletter got better.
We’re not done yet. In 2023 I want our newsletter to continue to improve.
If you want to see our newsletter yourself, you can subscribe here.
We removed “paper cut” problems from our product
Feature Upvote is six years old, which is like, say, 50 years in SaaS years. It is a mature and stable product. We completed most of the major stuff from our own feedback board long ago.
You know those products that look great, but actually using them feels like you are endlessly getting paper cuts? Those products you have to summon courage each time you need to use them?
So in 2022 our improvements mostly involved polishing existing features. We removed rough edges; we made small tweaks that make Feature Upvote easier to use; we found and fixed performance issues. We removed many “paper cut” problems.
I like these types of iterative improvements. They make for a great product our team can be proud of and our customers enjoy using.
You can see a list of all the improvements we’ve made in our changelog.
We gave Google Ads a big effort, then gave up on it
A couple of times in past years, I’ve dabbled in Google Ads. Because Google Ads is a confusing product that requires plenty of initial learning, and a ton of ongoing attention, I always stopped after a couple of months, scared that I was throwing money away.
In 2022, with a marketing person on our team, we gave Google Ads another shot. For six months, we created and honed our ad campaigns.
But we never really seemed to get good return on ad spend. We never could work out the perfect combination of ad copy, keyword targeting, and landing page optimisation.
So we stopped Google Ads for the foreseeable future.
I’m not too upset by this. I’d rather take the money we would have given to Google, and instead use it to sponsor grassroots podcasts, newsletters, and events.
(By the way, do you have a podcast, newsletter, or event that is seeking sponsorship? Then contact us. Especially if your podcast, newsletter, or event is quite new and you haven’t had a sponsor before – we love supporting people getting started.)
What’s up for 2023?
More of the same! We have a good team, and the company is running smoothly. We’ll continue our approach of making steady progress with a series of small changes, each hardly noticeable, but that add up to an ever-improving product. We’ll keep doing our best to keep things simple: simple for our customers, and simple for the people who contribute to your feedback boards.
Need a product ideas board? Give Feature Upvote a try!