Congratulations! You just realized you are running a software company.
You didn’t start that way, but one thing led to another, and now you have a small team of engineers.
They may be in-house or outsourced, but you are managing them anyway.
Software is tough. Engineers seem to drive you crazy.
It feels like you cannot get a straight answer out of them.
- “How long is this going to take?” → “It depends”
- “How many people do we need?” → “It depends”
- “Why is this feature not what I asked for?” → “You didn’t use the right words when you asked for it.”
When the feature finally ships to the customers, they find a bunch of problems.
Why didn’t the software team find those beforehand?
And let’s not talk about the outages.
You wonder why software is so hard? People engineer all sorts of other things with way fewer problems than software.
- Maybe you heard about agile or scrum and asked your team to estimate their work and get some damn predictability into the system.
- Maybe you went with OKRs and tried to set some high-level business metrics.
- Maybe you heard of DORA metrics and tried to figure out how to start fixing this.
But so far, little is working.
It’s pretty natural when all appears lost to try to quantify things.
If we can quantify the problem, we might be able to figure out the solution. Isn’t that the whole idea behind being data-driven?
Most of the time, though, metrics are not the answer. They are a good tool but usually applied way too early.
The first thing to change before reaching for metrics is your goal mindset.
Start by thinking about about what matters to you. Presumably, this is some variant of delivering value to customers via shipping code.
To optimize around doing that:
👉 Start with small changes on what are your biggest bottlenecks. Try to bite off something huge, and you will fail, losing the precious momentum you desperately need.
👉 Get consistent, incremental output. Consistency shows that the improvements you made are good and not just lucky.
When is it time to reach for metrics again?
- When you have solved your biggest, most obvious challenges.
- When you are stumped as to where your next problem lies.
- When all your potential investments are very expensive, and you need harder data.
Easier said than done?
Bluntly, yes.
You must allocate significant resources to this process and ensure you do the right things.
If you do not fix your bottlenecks, you are either starving your most precious resources or generating intermediate work piling up somewhere.
If this resonates with you and you could use help navigating it, ping me, and we can discuss your bottlenecks and how to eliminate them effectively.