The clearest way I can help you understand why story points aren’t a measure of time
For the purposes of this thought exercise, I want you to pretend that google maps and all other route planner software has been eradicated.
Say you’re going on a journey. You know that from point a to point b, it’s 125 miles. Now imagine I asked you straight off the bat, without any information other than the simple fact that it’s 125 miles away, how long will it take? You’ll most likely have a plethora of questions. What’s the mode of transport? What’s the speed at which we’re going to be travelling? What’s the congestion like at the time of travelling? Etc etc. Without knowing some of those things, the prediction you give for how long it take will be a wild guess at best, littered with assumptions. Now the stars might align and all those assumptions come true, stranger things have happened, and you actually took exactly how long you thought it would. You’d be quite the unicorn at that point.
But the reality of the fact is, that assumptions are very rarely accurate. So your initial estimate, which I like to call a BEG (Best Educated Guess) of a 125 mile journey might be somewhere around 2.5 hours, assuming 50 mph as an average travelling speed. Now what if you don’t have a working vehicle to use and have to rely on public transport? Suddenly all assumptions are out the window and that original BEG of 2.5 hours is probably going to be nowhere near enough. You have no idea whether there is a direct connection between the two points, or even how many modes of transport you’re going to need to take, or the wait time between each of the legs of your journey.
For simplicity sake, let’s say you do have a working vehicle and upon a first high-level look at the journey, about 50% of the journey will be motorway and the other half will be urban 30mph roads. You’re starting to build some confidence in your initial BEG of 2.5 hours now, and that’s great. As more assumptions are crystallised and confirmed you will develop more confidence. But, you’re still making the very big assumption that you’re going to be able to do the speed limit 100% of the time. What if there is road works on the motorway limiting you to 50mph? What if there is an accident and some lanes are closed or even worse a complete stand still? What if it’s your vehicle that breaks down? Is there a back up for that? Do you have breakdown recovery? What if there’s a diversion and you end up having to travel further? All of these things, are assumptions that cannot be confirmed ahead of time. When you set off on your 125 mile journey, you have an idea of how long it should roughly take. With every mile you travel, you reduce that cone of uncertainty down to narrow in on a more accurate ETA. The reality is, that until the end line is physically within your line of sight and you’re able to confirm the last of your assumptions, your ETA is only an educated guess. That’s not to say that something couldn’t go horribly wrong in the last 15 metres of the journey! But chances are very low and an extreme edge case.
Now imagine the 125 mile journey is your release. The team is the car and the roads you’re travelling on is the stories being delivered.
The same goes for story points. If an MVP release consists of 125 points worth of stories within it, being delivered by a team that typically delivers around 25 points per sprint (assuming 2 week sprints), then straight off the bat you know that your first BEG for a delivery date is in 10 weeks. There’s nothing wrong with having a date that you’re using as a guide for discussions as long as you make it clear and remember that it is only a guide and everything is subject to change.
After one sprint in, the team has now delivered 20 points of the 125. I’ve purposefully not used the adverb ‘only’ here, because that would entail that your BEG is then becoming a hard and fast deadline, something you need to avoid. Now that you’ve travelled 20 of your 125 points, your cone of uncertainty has begun to narrow down. Sprint after sprint you’ll adjust your understanding of the speed at which the team is delivering against the 125 points and be able to gain more confidence in when you’ll be delivering the last point. But there is a plethora of things that could cause the team to slow down or speed up.
Let’s compare them to our journey from point a to point b:
- Traffic jam = interdependencies
- Lane closure = reliance on a single body of knowledge
- Breakdown = sickness/attrition or the realisation that it’s not technically possible to achieve
- Diversion = scope creep
- Roadworks = up-skilling is required
- Speeding = Flow aka uninterrupted high-performance
Each one of these things is something that could cause deviation from your original BEG, and should be considered in all discussions for when your delivery date will be.
As you continually travel the same road you get a good understanding of what kind of patterns of deviation you encounter and will get very good at predicting how long it’ll take you to do a particular journey. The same goes for stories. If you’re tackling a piece of work similar to something you’ve done before, on the whole the size of it will still be roughly the same (potentially there is less doubt and risk meaning it could be slightly smaller), but the team will be more versed in the challenges that they’re going to encounter and whether it will slow them down or not.
I talked about something similar in my other article, Software Delivery is like Wizards, for further reading check that out.
I am available to give talks in England on how to build a culture and environment that breeds productivity and fulfilment through effective leadership. Connect with me on LinkedIn and let’s talk some more: https://www.linkedin.com/in/martyn-puddephatt-14b08757/
