I was writing a new feature announcement email, and found myself describing this new feature in 7 (seven!) easy steps. While writing this email, wondered why I was exposing my users to some of the implementation details. So I sat back for a moment, opened my code editor, changed my user interface, hiding all the internal stuff and deployed the change. Then I sent an updating email and enjoyed for a brief moment a feeling of achievement. Then, it struck me: product management for a SaaS product is tough and I still have a lot to learn.
Here is my personal story of being product manager (and co-founder, developer, and chief customer happiness) of DenBot by Sirris.
Over a year ago, my colleague Omar Mohout and I started toying with twitter automation. We wanted to see how far one could get by automatically retweeting and favoriting tweets with specific keywords. Our software version 0.1 was a simple, single purpose built piece of software Omar commissioned through one of those freelance market places. I adapted the code and we had it running on free hosting. It worked fine for the both of us, nothing more, nothing less.
After some time, we realised that Twitter automation was really valuable for us. We started talking about it with others and before we knew it, we had people asking if they could use our tool too. I deliberately use the word tool here, because that was what it was: a utility, it didn’t look nice, it was single tenant, harsh and not meant to be used by anybody else but Omar and myself.
Tools transformed into SaaS product
So Omar and I agreed to transform this tool into a real SaaS product, that we would offer to customers. Or to be more precise, we would explore how far we could get in this process. I would take up product development, support and a bit of marketing, while Omar would be the 'evangelist' and first (power) user. After all, how hard could it be? We had a working prototype, we just had to make it a bit more user-friendly, add multi-user support and that would be it...
Well, as it turned out, it was a lot harder. The technical side was the least of our worries. We had and we still have the occasional hiccup: processing over a million of tweets in real time caused a few challenges of its own. But frankly, these issues were easy to fix. There’s so much information and code out there to help.
No, the really hard part turned out to be understanding the users. We thought we knew, as we were using our own product, our twitter bot, every day. Still, the gap in perception between me, as the DenBot product manager and developer, and the users is large.
Bridging the gap
I should have known software product management is tough: for seven years now, I coach software entrepreneurs and often, product management has been one of the core areas that needed improvement. Product managers need to balance a gazillion requests, from different stakeholders: the sales people have requests for new features, developers want to improve code and architecture, users have feedback on the product, support wants better documentation, etc. In the midst of this turmoil, it is the product manager and his team that guard the vision of the product like a watch dog.
How I implement my role as product manager
There are two things that guide me in my works as product manager for DenBot: 'Product Vision' and 'Feedback'. Let’s zoom in on each of them:
On a higher level, I know exactly what DenBot as a product needs to be: a SaaS that offers its customers the possibility to intelligently automate the animation of their Twitter account. As such DenBot should be a multitenant and scalable self-service online platform, easily supported by a small yet highly engaged team. This vision gives me focus and it gives me the courage and backing to say "no" to users when they ask for advanced Twitter analytics. I could do this, but that would distract me and, more importantly, the users from the core feature of DenBot: the intelligent automation of Twitter accounts.
This vision shows me the big picture and gives me guidance. Feedback helps me to get the details right. It is by actively looking for feedback that I (try to) figure out if the translation of my product vision into a tangible product is okay. The main tools I use for feedback are:
- Talking to users: I take every opportunity to talk to users, learn from them how they use DenBot, how they perceive the product and what it can do for them. It’s really humbling to learn that things are not so obvious for users as it is for me, despite the fact that I thought I built DenBot with usability in mind.
- Analytics: as the product manager of a SaaS product, I have the luxury to observe users as they use my product. Tools such as Inspectlet let me record videos of sessions of users with just a snippet of Java script. Again, it is very sobering to see how users hesitate to fill in things, despite the help texts, despite the, in my opinion, obvious user interface. Analytics give me insight in what features are being used and how often, who’s using what . Analytics allow me to learn what (might) work and what not from a user’s point of view.
So one day, I hope to be better at product management, but frankly, I think, as a product manager of a SaaS company, you’ll have to keep assuming you're no good. You’ll have to keep working on bridging that gap between what you think customers need and how customers really perceive and value your product.