top of page
  • Writer's pictureSavita Ferguson

How do you know what to try?

Updated: Jun 28, 2021

Some years ago I worked with a team that was struggling to ship. It was a small team of 7 team members and they didn't really have any dependencies on other teams, so this wasn't a problem of scale. But I suggested using a release train cadence - every Thursday, everything that was ready to go out, went out. If it wasn't yet approved for release by the business it would go out behind a feature tag, but it went out. This worked, the team that had been struggling to release finally found their stride in regular releases.

Fast forward a few years, someone who knew what I did on that team suggested doing a similar release train concept for a different team. But I thought this was the absolute wrong choice. Why?

Because they didn't struggle with the same problem.

Every single situation you find yourself in as an agile coach is entirely unique; this is why just applying an agile framework rarely (never?) works - it always has to be tweaked for the specific constraints of the team you are working with.

Because I advocated strongly for this small team to do release trains, this person mistakenly assumed I was a fan of them; I'm indifferent. The new team he suggested applying it to practiced continuous delivery and shipped several times a day. I would never suggest asking them to release with a slower cadence, only on a Thursday. Different situations call for you to deploy different strategies - so how exactly do you know what to try?

Understand the problem - It sounds fairly obvious but it actually takes a lot of conscious effort. For example - think about the complexity involved in writing a survey to understand employee opinions - it actually is a specialist skill to craft a survey that gives results that are useful. Something as simple as the word choice when writing a question can unwittingly influence the respondent. My general advice here would be to make sure you acquire some of these skills on how to discuss things in a way that gets you information as close to the truth as possible. Seek multiple perspectives.

Understand your options - There are so many books, blogs, videos etc. with ideas for how to resolve a problem and google has made it so easy to find these resources. Do your research! Find these ideas and use them!

Follow the connected links to other ideas that are somewhat but not exactly related to the problem you're trying to resolve. If there's a book mentioned, go read that book or follow the author on twitter.

Find communities. Networks of individuals who think about the same problems is invaluable. Make sure you have individuals who you can bounce ideas of; who can help you recognize bad ideas for what they are and help you tweak promising ideas into good ones. Use local resources at your workplace but also remember how connected the world is and find individuals via virtual meetups if you think that would be useful to you.

And just a plug here as a reminder that sometimes completely unrelated ideas can provide such creative fodder for problem solving. Yes you might be primarily operating in an IT setting, but you'd be surprised how useful ideas from other fields can be. Completely random example, a long time ago I read that in Pret a Manger, interviews for baristas were conducted as an onsite trial for a week. The idea is, the space behind the counter is so small, the team members need to be in a kind of dancing harmony for it to work well. I sometimes visualize the team members I work with doing this dance; not necessarily co-located but the way they pass information between each other; the orders, the status of things being built, whether someone was a bit more clumsy or it is all smooth.

Understand the possible implications of your options - I sometimes like to draw a matrix of my options: how difficult it is to do vs how longstanding will the benefit be. Just thinking about the options in this way helps me narrow down which ones are worth exploring and which others I'll just keep in my back pocket.

Also, I fully recommend thinking about the human implication of any choices. Who are the stakeholders here - anyone who affects or will be affected by the choice. I find learning more about ideas from the field of cognitive behavior and psychology very very useful in an office setting.

Try something - Data collection, data analysis, research can all take a very long time. How do you know the point of optimally stopping where you've got the right solution. You don't. But you need to always remember there is a cost of indecision.

I have an understanding with my boss that. I am sometimes going to screw up and make mistakes, if I don't, then it means I'm not trying hard enough. As an example, I was working with a team that was struggling to jell as a team. I thought it would be a good idea to have them work closely with a very tight knit team, just as an example (reminder) of how things could be different. The idea was that this would inspire them to do things differently within their own team. The reality though was that one of the team members said it made him realize how unhappy he was in his team. Instead of bringing the team closer, I accidentally pulled them a bit further apart.

Sometimes you're going to make the wrong choice. But most choices are recoverable. And it is often better to choose and act deliberately, than to just do nothing and hope for the best.

Good luck!


Algorithms to Live by: The computer science of human decisions by Brian Christian and Tom Griffiths

3 views0 comments

Recent Posts

See All


Post: Blog2_Post
bottom of page