Ken Furlong

One of the biggest problems (if not the biggest problem) we have on our team is technical debt that has accumulated over the last 4-5 years. Fortunately, the team and our Product Owners understand the problem of technical debt in general and recognize it in our case.
We’ve started taking small steps to reduce our amount of debt. I’d love feedback and suggestions about what we’re doing and anything that anyone else has found helpful.

The first thing we’ve done is resolved to not take on any additional debt intentionally. Of course, almost anything we implement will eventually become technical debt if it is allowed to collect dust long enough or if circumstances change – but the key point is that it wasn’t debt when we implemented it. Our hope is that the rate at which technical assets become technical debt will be such that we will be able to keep up with regular refactoring.

The second thing is that our POs have made it clear that they are willing to give us the time to pay off technical debt – but the burden is on the team to identify, flag up, and explain the technical debt to the PO so that it can be properly prioritized in our backlog.

The third thing is the team now has a weekly 30 minute meeting to discuss technical debt. We don’t have a firm agenda, but the discussion usually centers around a few points:

• Are there any pieces of technical debt that we would like to discuss (presumably because we haven’t discussed them in this forum before)?
• How costly would it be to pay down this piece of debt? (We estimate this using the XS, S, M, L XL scale.)
• How costly in the interest on this debt? That is, how much pain is it causing? (We estimate this using a yellow, orange, red scale. I’ll explain why in a minute).
• How should we begin the process of paying down this debt? Is this something we can “just fix” with a little effort on the side? Is this something that we should write up a user story and request our PO add to the backlog? Should we keep it in our back pocket for a hack-a-thon project?
• Who is on point for this piece of debt? That is, who is going to “just fix it” or write the user story or keep it on their own hack-a-thon to-do list?

The fourth thing is that we’re now maintaining a technical debt board on a wall near our sprint backlog. We wanted a visual representation and reminder of our team’s biggest problem. It will hopefully help us stay focused on it, not let us forget about any given piece of technical debt, and help us track and encourage progress (a very important facet, in my opinion).

This is why we estimated cost using size but impact using color – we can visually represent each piece of technical debt using a piece of paper, not card, post-it note, etc. of the appropriate color and have a board where someone can assess all of the critical information at a glance (example). If we had just used Fibonacci numbers for each scale and written them in the corners of cards or something, it would be much harder to get a sense for the whole situation.

So far, we’ve identified two pieces of technical debt that we can “just fix” in our spare time and the fixes are in progress. We’ve also begun working on writing the stories necessary to eliminate another debt. Hopefully, we’ll be able to keep up this momentum, increasing our velocity and quality along the way.

Any tips from those who have gone before would be greatly appreciated!

Hide Sites

Since converting to Scrum, my team has been in the practice of planning our capacity for a sprint in terms of ideal hours.  We had a fairly simple spreadsheet where we’d enter the number of vacation days each team member was planning and taking and their estimated “overhead” percentage (all of the time spent in meetings, handling random things that come up, etc.).   During our Sprint Planning meetings, we would then estimate all of the tasks for the stories in terms of ideal hours – how long we expected that task to take assuming zero distractions and interruptions.  We then took on as many stories as we had enough ideal hours for.

Over time, I became less and less satisfied with this way of planning capacity.  In general, it didn’t seem to add much value to the process and increased the length of the planning meeting.  Additionally, because we track the progress of a sprint using the number of hours of work remaining, the team had to continuously update both our task board and our online tool (whether Rally, ScrumWorks, etc.) or others wouldn’t have a good sense of how the sprint was going.

At best, this represented time (albeit, not a ton, but still enough that it hurt) not spent doing actual work.  At worst, it was a complete waste since there were usually caveats associated with the hours as they are presented on the board or in the tool, e.g., “well we’re way over in terms of hours we spent on this task, but we realized that all the work we did will save us time on the next 5 tasks so it’s basically a wash” or “we’re going to leave this task at 6, but we might lower it to 1 shortly depending on how something turns out”.

One could respond that those types of things can and should be tracked in a tool and the problem is not that we were using ideal hours, it was that we were being lax in updating the tool and, by extension, the rest of the team and stakeholders.  While this was initially my thought, I came to disagree for the following reasons:

1. It seemed odd that we were estimating work in terms of a fictional unit – the ideal hour.  Since there is very rarely an extended period of time during which someone really doesn’t have any distractions and is free to focus on a single task, I don’t understand why we ask them to imagine how long a task would take under those conditions.  Granted, it makes the math easier, but that doesn’t make the estimate any better and might actually make it worse.
2. We limited the granularity of ideal hours to whole hours.  Even if ideal hours were a real unit, limiting their granularity means limiting their accuracy and usefulness.  Granted, estimating in whole hour blocks is faster and easier, but the very mechanism that makes it such also severely limits its usefulness – especially when there is disagreement within the team about how long something is going to take and we just settle with the average.
3. In my experience, estimating in ideal hours didn’t help us.  There were times when we had to go to the Product Owner and say that we couldn’t get all of the stories that were tentatively put on the sprint backlog done in time (because the number of ideal hours needed was higher than our capacity) – but those were identically the times when the number of story points on the sprint backlog exceeded our velocity and/or where we had huge 8 point stories which we all later agreed should have been multiple stories whose points would have added up to more than 8.

For a while, I didn’t have a good suggestion as to how to do away with ideal hours.  Then I saw this presentation on the ScrumAlliance website.  I’m going to take the liberty of paraphrasing the author: if we’re really doing our story point estimates well, and we’re always trying to break things into smaller stories so that story sizes are as uniform as possible, and we’ve got some historical data to tell us what our velocity is, why don’t we just use that to figure out how much to put into a sprint and save ourselves the trouble of estimating tasks?

Under such a scheme, we would still task things out so that we could uncover any gotchas and get a general consensus in the team as to what needs to be done.  We just wouldn’t estimate hours for tasks and then track progress in terms of hours remaining.  Granted, it might turn out that we sign up for an amount of work that causes us to either end early, have to work a little extra, or miss a story, but we run that risk with ideal hours too – we just spend more time doing it.

In terms of tracking the progress of a sprint, the author suggests having a burndown of tasks rather than hours – which is obviously less quantifiable but perhaps no less valuable.  Our current burndown uses hours, but it really just gives the illusion that we know exactly how many hours are remaining.  Not having an ideal hour burndown just means we don’t have that illusion anymore.  As the author points out, precision doesn’t equal accuracy and accuracy is what we’re really after.

Lastly, there was the issue of how to adjust the amount of work we pull into a sprint when we know that someone will be on vacation.  Ideal hours gives us a nice way to do this because we just subtract the appropriate number of hours from our capacity.  That really isn’t that accurate, though, because any given day in a two week period might be very different than any other day in terms of how much someone is able to focus on direct work.  Treating all days as identical in terms of capacity is again mathematically easier but perhaps no more accurate.  We could probably do just as well by “manually” adjusting the velocity down a few story points based on gut feelings.  Again, precision doesn’t equal accuracy.

When I presented these ideas to the team, we all decided that it was worth trying – after all, even if we crash and burn, we’ve only lost two weeks.  That was 2 sprints ago and we all seem pretty happy not using ideal hours.  Our task board is just the same except we don’t put hours on our tasks.  Our online tool is the same – we just assign every task 1 estimated hour.  Our burndown chart thus shows the burndown of tasks.  During planning, we are extra conscious of task breakdowns and try to make all tasks as uniform a size as reasonably possible.

All in all, I’m very happy we’ve moved away from ideal hours and are relying more on our velocity and “gut checks” to know how much work to pull into a sprint.  I highly recommend trying it.

Hide Sites

Jack Milunsky has a brief post up over at AgileSoftwareDevelopment.com wherein he discusses the issue of switching stories mid sprint.  One of his points that I’d like to draw attention to is his response to the criticism that a team can be the slave of the process, i.e, too rigid in following Scrum mechanics and not willing to change a sprint mid stride because something urgent comes up.  As he says “Well you’re either a slave to the process or the team is a slave to any chicken in the company who shouts the loudest.”

That is a fantastic point, in my opinion.  It is much like the oft repeated response when something goes wrong while you’re doing scrum: “would this have been any different if we were doing waterfall?”

Hide Sites