Lesson 1 from the apocalypse: exponential growth

This article is a translation from the original French post on Arolla‘s blog.

Since he worked remotely full time, Pluto Krath had set up an office space in his kitchen. His headset jammed on his lock-down hair cut, he was facilitating the planning for the sprint to come.

— I know we have such a pile of bugs. But we promised the boss this feature, we have to release it in 2 weeks.
— …
— Yes, I confirm we are blocked for the next 3 sprints. We committed, we need to deliver. We’ll check out on bugs later.
— …
— OK, do whatever you can to work around them. And take technical debt if needed.
— …

At the end of the meeting, Pluto closed his laptop. He really needed to avoid having conflicts every time with the team. He had to find a way to fix those bugs. But was it his fault if the team always committed to deliver too early? if estimates always fell behind?

— It’s because you know exponential growth goes faster, said a voice from the fridge.
— What? No! Not here!

On the sharknado poster he had saved from the office, Tara Reid was missing. She closed the fridge, holding a soda.

Tara Reid

— I’m your conscience from the world after. I replaced David Hasselhoff.

Pluto almost fell from his swedish chair, holding on to the tablecloth.

— Why are you talking about exponential growth?
— Bugs are like any other virus. They spread. Exponential growth. Were you here during the last months?
— You’re comparing bugs and a pandemic. I never saw anyone die from our bugs.
— I’m talking about models. Bugs cause bugs. They prevent you from coding, testing, debugging, fixing, getting relevant feedback from users. They decrease your standards. The more bugs you have, the more bugs you get. Exponential growth.
— OK, let’s consider it’s a virus. How do you stop an exponential growth? You can see that every country is in the same dead-end. We destroy the economy and wait for the vaccine. And I don’t know any vaccine against bugs.
— Oh for the economy you’re probably right. Everyone depends on everyone, and every country must adopt the same measures sooner or later. But for health, that’s a different story. Compare numbers between countries that had never experienced SARS, and countries with a 0-case policy: 1800 deaths in South Korea, 100000 in France.
— They say it’s easier to protect an island, but I never bought that explanation for a 50 fold ratio, you’re right.
— Now dig into your memories from the beginning of the epidemic. Epidemiologists knew we were about to deal with an exponential growth.

— I was struck that no one saw it, and that people in charge didn’t react either… willingly or not, by the way.
We are not wired to understand exponential growth. We see our current situation without remembering how we got here. And we can’t extrapolate the future from here.
— Especially when we don’t want to see.
— 😁
— Now that I’m understanding how you work, I suppose I need to search my toolbox and understand what’s going on. Probably in lean, as usual… Oh! 0-defect? andon? jidoka?
— 😁
— We stop the chain at the 1st defect, we take the time to understand the situation on site, and we fix it. I see. Anyway I’m stuck. I can’t do anything about bugs, I need to deliver features.
— 🤦‍♀️
— Can you please keep your head vertical and stop speaking with emojis?
— Listen to yourself. You want to deliver only features because you committed on arbitrary dates based on estimates. Estimates are always wrong, that’s another cognitive bias. Nobody demanded those features for a particular date. Now, you can’t fix bugs because you can’t deliver features because you have bugs. You’re about to add bugs to deliver features faster than your product can bear because of bugs. Everything would be faster if the team fixed bugs on the fly.
— And as the delivery rate already included a continuous flow of bug fixes, we could estimate features based on what we delivered yesterday.
— Figure it out yourself, I don’t care about your spreadsheets. What matters is delivering value sustainably. Not to make promises you don’t want to keep. Oh! And stop with your technical debt, you don’t understand what it means.
— OK I’ll see if I can negotiate the delivery dates in order to schedule bug fixes now. Thanks Tara.

Pluto headed for the door, shut down the light, looked at his poster with a smile, knowing he would see Tara on it. He startled when he saw her sitting on his chair.

— Do you think we’re done? We’re talking about exponential growth. It’s not only about bugs.
— What? Features?

He turned the kitchen light back on.

— Features are a consequence of it. Come on! Search a little. What else does pile up?
— …
— Quality! Every tool, standard, successful deployment, code deletion, help you deliver users what they need. It gives the time you need to understand what they need, and to sharpen your tools.
— Do you expect me to cut it as well?
— On the contrary. You want it to grow exponentially. Do you know how much you earn at the end of the year if you grow by 1% every day?
— 1.01 to the power of 365, that’s a 37 fold increase.
— There you go, a real calculator. I see why you’re a high potential. While you’re on fire, can you tell me what makes product quality?
— The vital forces of our company, men and women making the heart and soul of our great family, of course.
— Good, my marketing slide. Wisdom flowing between the team’s brains. And guess what?
— That wisdom is also exponential. It allows you to find the time to learn, it sustains itself.
— That’s it. Get trained, share your knowledge, take the time to understand. That’s instant return on investment. And now, tell me what strategies you adopt to cut bad exponential growth and reinforce the ones you need?
— Well, we fight for each bug prioritized, when they don’t make us stop everything else when we got ridiculed in the media. We don’t take the time to get trained or think and understand. And we stop investment projects as soon as we get any more or less tangible result.
— Even when you look at bugs from every angle, label them, sort them, you don’t change anything about their proliferation. You have to fight with your managers to improve your efficiency and skills. And the worst is, I’m sure you chose on your own to be slowed down by your hierarchy.
— Now that you mention it, it’s possible.
— Well I have 2 good news for you. Your competitors all do the same, and you have the perfect opportunity to make a difference. Do what needs to be done, you don’t need approval from your hierarchy. Even better: let the team do what needs to be done.
— OK let me think about that. Cut dampeners at their roots, foster accelerators, continuously.

Pluto got lost in his thoughts:

— I suppose we’ll try to deliver smaller things, more complete, with a dose of WIP limit and jidoka. I’ll try to be clearer on encouraging learning. What if I shared links of conferences to watch on work hours? What if I deleted that “assigned to” field in tickets, to encourage collaboration? We’ll talk about these tomorrow.

This time, whether Tara stayed or not, he decided to go to his family in the living room. He turned back to shut the light down, and it suddenly clicked.

— By the way, how do I know when something will grow exponentially?

Tara was gone. The poster was back to normal. He supposed some other actor from sharknado would come back some day to explain the rest of the story. If only it could be Dolph Lundgren.

One comment

  1. Pingback: Lesson 2 from the apocalypse – expertise | AAAgile

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s