Our journey to customer-centric feature management

Feature management vs product analytics

We’ve always been an opinionated, mission-driven company. Since our founding in 2021, our mission has been to empower product teams to build better features.

In practice, this means getting really good at not only shipping product updates but quickly reviewing them and then iterating until customers love them. When R&D works this way, it’s transformative for long-term product outcomes.

But for better or for worse, we never picked a product category. Our priority was our mission. If that meant combining data and features from different product categories, so be it.

Initially, most of our data points were metrics-based which understandably made customers perceive us as a product analytics tool. Quite ironic since Bucket was partly created in rebellion to product analytics! 

Product analytics tools have been around for more than a decade. During this period, I’d claim that most B2B R&D teams haven’t truly become data-driven or customer-centric. We’ve experienced this ourselves as former product owners and have continuously heard the same things from our peers. Giving product teams access to engagement metrics is one thing — making those metrics actionable and actually impactful is another.

The reason? First of all, they’re just metrics. It doesn’t help you uncover the why behind them. Second, to get those metrics, each team still has to do event aggregations, custom engagement definitions, bespoke dashboarding, and manual reporting for each release. Thirdly, they’re horizontal products designed to handle many use cases but aren’t focused on doing any specific ones particularly well without a ton of customization. 

Bucket was never here to replace product analytics. Bucket was created because the problem we solve is far too big and important (tech companies spend 50% of their costs on R&D) to be handled by something that isn’t specifically built for it. The solution goes far beyond engagement metrics you get product analytics. 

To answer the why behind the metrics, you need to talk to customers. That’s why we added event-based feature surveys to automate this part of the feedback loop. But even after adding qualitative insights, customers still kept placing us in the product analytics category. 

Initially, we didn’t mind as we thought eventually it would become clear we were so much more than product analytics as we executed our roadmap. 

But that never happened. And even if it would have, if we weren’t product analytics, what were we?

We should have known better. 

If a company does not deliberately position their product in a market, one of 2 things will happen:

1. The team will collectively use a “default” or assumed market position.

2. Customers will decide what you are using the clues they can most easily assemble.

We did choose so customers continued to place us in the product analytics category. That meant trouble. We’d do a product demo with a product team member, it’d go well, but then we’d get hit with: “By the way, other teams are using product analytics for use cases A, B, and C. Can you help with those? If not, I can’t get those teams to replace our current product analytics tool with yours”.

Obviously, we had a positioning problem. For six months, we struggled with this issue. 

Do we just keep going? That didn’t seem like an option. 

Or are we product analytics with qualitative insights? Qualitative insights with engagement metrics? e Feature management with customer feedback? 

This March, we drew a line in the sand and picked a product category.

We’re Feature Management. Actually, we’re more than that. We’re Customer-Centric B2B Feature Management.

For most, feature management means feature flagging, which is owned by engineering. To us, feature management is so much more. Feature flagging is more of a necessary evil than where the value is. 

The product and engineering functions are more and more starting to converge, particularly at fast-moving tech companies. Companies want empowered product teams, they’re hiring product engineers and they’re optimizing for iteration velocity. This means product teams, as a whole, are now using feature management. 

Whenever product teams roll out an update, they should be ready to act on customer feedback and iterate accordingly. A successful feature rollout doesn’t end with stable technical performance, it ends when customers love it.

Release, review, and iterate isn’t a new concept, but it’s much easier to talk about than to do in practice.

To actually do it, product teams need leading indicators, both qualitative and quantitative, and they need them to surface automatically. Bucket already provides those indicators today and now we’re adding feature flagging so you can roll out features based on customer satisfaction.  

We’re staying super focused on weeks 0 to 12 for every product update. Feature management is the perfect category to do this in. There’s a place for it too. We’re bridging the gap between feature deployment and product analytics. Doing so is tremendously impactful for any tech company’s R&D ROI and customer satisfaction.

Check out our all-new website and stay tuned for more leading indicators 🍿