Stop suffering from 'Goldfish Memory' at the Sprint Review
The previous Sprint is least interesting to talk about
In the movie 50 First Dates, Adam Sandler falls in love with Drew Barrymore. There’s just a slight hitch — she has amnesia and loses all the new memories she forms when the day ends. This prevents her from falling in love with Adam Sandler for longer than 24 hours.
As a result, every day, Adam Sandler has to start over from scratch and try to win her over again and again, hence the title.
In my experience, many Scrum Teams suffer from the same phenomenon as in that movie — what I’d like to call Goldfish Memory*.
Goldfishes only have a memory that spans a few seconds. The teams act like fishes swimming around in circles every Sprint over and over again to ultimately end up at the same point.
Except that the fishbowl they are circling isn’t real. Their glass container is an imaginary and self-imposed barrier.
Endlessly revisiting the last thing you did is ineffective
Now imagine a bodybuilder who only reflects on the last time she went to the gym and doesn’t reflect on meeting her goals. A Formula 1 driver who only cares about the latest lap time he put on the clock. A stock trader who only looks at his stock when the closing bell rings.
It’s hard to imagine these professionals being successful at what they’re trying to achieve without considering the big picture. However, this is exactly how many Scrum Teams operate.
At the Sprint Review, the Scrum Team and stakeholders inspect the Product Increment. Except they are not inspecting the Product Increment, but only a tiny part of it.
Many Scrum Teams spend most of their time talking about what they did in the last Sprint, mistakingly believing this is the whole Product Increment. I’m talking to you, teams that spend the majority of the Sprint Review giving demos of what they delivered in the past Sprint.
That’s like someone who tries to make a delicious hamburger and only talks about the last piece of lettuce they added on top.
What you did in the last Sprint often constitutes less than 5% of the Product Increment, assuming you’ve been working on the Product for more than a year with two-week Sprints.
In fact, as more Sprints go by, this number will drop even further. The only situation where most of your Product Increment will have been delivered in the past Sprint is when you’re just starting.
The Product Increment is the whole product, including what you made the previous Sprint and all the Sprints before. Here’s how it is phrased in the Scrum Guide:
“Each Increment is additive to all prior Increments and thoroughly verified, ensuring that all Increments work together.” — Scrum Guide 2020
Like a hamburger is more than the last piece of lettuce you’ve added on top, your Product isn’t just the work you did in the last Sprint. You need to look at the whole Product. It is essential to consider the whole hamburger to see how you’re doing and not the final touch of adding that lettuce leaf.
Plus it’s not about what you’re adding to the hamburger, but how it is experienced by the person doing the tasting.
‘I know what you did last Sprint’ isn’t good enough
As Paddy Corry has written before, “Done doesn’t mean it is done, it means ready for feedback.” The fact you’ve deployed something to production is only the first step on your journey.
Instead of suffering from Sprint myopia and only talking about what you did in the last Sprint, here are some questions you could be asking instead:
How is our Product doing? Are we making progress on our Product Goal? Are the features we’ve built helping the customers to meet their goals?
Are there features we should be ditching or reworking?
Is the Product Goal a concrete step towards our Product Vision?
Is our Product Vision still accurate, or should we update it based on the latest insights we’ve obtained?
Every Sprint, you collect more data and get more feedback on how your Product is doing. Don’t waste the opportunity to reflect at the Sprint Review by only considering what you’ve worked on recently. In fact, try to deploy new Features to production already during the Sprint to collect real-world data on how it is being used. Observing how it’s being used is much better than just talking about how you think people will use it.
Instead of only showing how it works, show initial data on how it is performing and the plan you have in place to collect more data to gain confidence whether the feature is performing or not.
Every feature is innocent of delivering value until we can prove its guilt. Expect your features to suck until you have evidence to hand over it is otherwise.
Only discussing the past Sprint is a wasted opportunity to inspect and adapt how your Product is really performing.
At best, a Sprint is only a small step in the right direction. To see how you’re doing, you need to consider the whole journey and not just the last step you’ve made.
*Goldfish Memory is a lie, scientists now believe they have a memory that spans many months, not just a few seconds.
** In this article I often talked about ‘the previous Sprint’, but this is meant colloquially. In reality, it’s still the current Sprint until the next one starts. I assumed this would be confusing for readers. Therefore I settled for calling it the previous or last Sprint.