It’s Friday night and you are exhausted from another week, fraught with the pressure and stresses of life. The thought of preparing a home-cooked meal for your family seems unbearable, and another fast casual meal feels like it doesn’t properly award your week. Tonight, is the night you take the family out for a nice sit-down dinner. After all, you’ve earned it.
So, you pack up the family and head to the local sushi joint, the whole way anticipating the mouth-watering kaleidoscope of deliciousness that awaits. You arrive, park in what feels like the outer reaches of a dystopian hellscape because all the good spots had been taken by the early-bird crowd, and you assemble your caravan ready to embark upon the long journey to the front door. As you carefully maneuver through the maze of minivans and electric vehicles, your patience is repeatedly tested by the combination of incessant bickering and the zombie-like meandering of your fellow journey mates, who appear more concerned about conquering the level they are on in Roblox than they do about avoiding a human-on-car collision.
You finally arrive at your destination, and you triumphantly open the front door with the enthusiasm of a Roman warrior slaying a fabled beast. Victory is yours.
As you provide your family name to the host, they tell you it will only be ten minutes before they can seat you. Phew, you think to yourself. I can handle that. I’ve come this far; waiting another 10 minutes pales in comparison to the quest you just survived.
But then 15 minutes pass… then 20… then 25. You are not convinced you can make it to 30 without perishing from famine, but you persist. The tension is mounting amongst your tribe, and you’re afraidful of what lies ahead if you aren’t seated. A mutiny might be on your hands if the situation isn’t de-escalated quickly.
You travel steadfastly to the host quarters and inquire about your accommodations. They tell you that the tables in front of you are taking longer than expected, and because of their staff shortages, they aren’t able to seat as many tables as usual, but that it should only be another 10 minutes. What a mysterious realm you’ve entered where all wait times are aligned to magical ten-minute durations. It’s as though you’ve discovered the nexus of where time meets space, and a wormhole will open any minute to transport you to your table.
You return to your family, head hanging from disappointment, preparing to convey the news. It is received better than expected, but only because your family has been incapacitated from starvation, lying on the bench as though their spirit had been exsanguinated from their bodies.
By the time they finally summon you and your family, despite your ravenous hunger, your experience has already been sullied beyond reparation, and your Friday night has been tarnished. If only they had better managed your expectations up front.
The Point of the Story
At this point you might be wondering to yourself what any of this has to do with product development, but I think this anecdote highlights one of the biggest problems I encounter on a regular basis: diminishing the importance of estimation, and falsely maligning it to be an entirely negative practice.
The agile community often vilifies the estimation process as being non-essential, or even as a counterproductive force towards outcome-focused, iterative delivery. It’s often referred to as being a primary culprit for culture-rot by forcing a date-oriented framework onto what should be self-organizing, autonomous teams whose focus should be on outcomes, not timelines.
While there is rationale behind this sentiment, I find it to be somewhat misguided. Timelines are a reality for any organization so that they can adequately map out the growth of their business. There is no avoiding this, and in fact doing so is counterintuitive to almost everything else we do in our lives.
We don’t set off on a road trip without first gaining an understanding of the expected time of arrival. We don’t build a house without understanding how many months it will take from ground-breaking to key handoff. We don’t even go to the grocery store without thinking about how long you’ll be there. Estimating duration is a natural part of almost everything we do. It’s what helps us plan our lives around these events.
I don’t think agile teams would dispute any of this, and I don’t think it’s the estimation process itself that is ever in question. The true cause of the consternation over providing estimates is really the result of unnecessary pressure put onto teams to meet the timelines born from these estimates.
That pressure that teams feel from the dates that they’ve produced doesn’t result from the estimates themselves. The inherent process of determining the relative size of work is not capable of inducing that type of gravity. The pressure actually stems from the mismanagement of the expectations associated with those dates.
Just think back to the story I told in the opening. How much different would that situation have felt if the host would have provided a much more accurate estimate from the start? Or once they learned that the 10-minute estimate was off by a factor of four, what if they provided better transparency in the form of proactive updates to keep expectations better aligned? How might that have diminished the impact of the lengthy wait time?
The point being that the 10-minute estimate in and of itself isn’t what caused the tension; it was the mismanaged expectations. This is a completely resolvable predicament.
How to Avoid the Pressure
There are a few things an agile team can do to sidestep the anxiety of living in a date-driven environment.
First and foremost, focus on becoming better estimators. By improving your predictability, you greatly reduce the risk of setting bad expectations from the start. That’s the most surefire way to reduce any potential misalignment. This is obviously easier with features that are well defined and represent more immediate goals. These live on the pointed tip of the cone of uncertainty. If you aren’t good at estimating these, then you have some serious work to do. Start here and focus on improving your team delivery metrics until you are able to consistently produce a sprint plan that falls within an acceptable threshold.
I am indifferent to the approach you use to estimate. If you prefer to focus solely on relative determinations of complexity, fine by me. If you find value in trying to estimate work duration, all good; do that. I think there are many methodologies that can be applied to achieve consistent results. Find what works best for your team, and your organization and focus on improving that.
Once you have developed predictability within your individual iterations, you can now focus on moving up the cone, into the opening mouth of added uncertainty, while trying to assemble a predictable quarterly plan. This is where things get more difficult. As you add uncertainty, your ability to produce a predictable plan inherently diminishes. This is ok. You just need to account for this uncertainty in your estimating framework. Bring it up to a level.
If you equate story points to workdays as you define your sprint plan, bring that up a level and try using total number of stories delivered as your measurement for throughput. You’d be surprised how consistent this can be over time. If your story points are simply a measurement of complexity, apply that same methodology to producing your quarterly plan. Play with this and fine-tune it until you reach an acceptable level of predictability.
Again, going back to the sushi restaurant story, had the host better understood their operating model and the influencing conditions, they might have been able to foresee the delay and better predict a wait time up front.
The point I’m trying to drive home is that you don’t have to follow a prescriptive model to achieve your intended results. If you are being disciplined and using a data-based model to drive predictability, you will naturally become better estimators. Don’t be afraid of this process. Make it a core part of your team’s DNA. Become an expert in your operating model and strive to become the best estimators possible.
The second thing you can do to reduce pressure in a date-driven environment is to better understand your leading indicators, ie identifying patterns that can highlight an early impact to your schedule commitment, and then become comfortable communicating risk early and often. The sooner you can identify these impacts, the better.
Be radically transparent, and proactively communicate changes to schedule. You’d be shocked by how this takes pressure out of the situation, while opening a wide array of options. It allows leaders to make better decisions because the earlier they are made aware, the more options there are available to them. Deferring features, for example, is much easier to do at this point because there likely haven’t been expectations set with external customers. It’s far likelier that the impacts from this decision are internally contained.
What you want to avoid is being unaware and uncommunicative like the host at the sushi restaurant. Be open and honest, and try to maintain the integrity of your estimates, but as soon as it becomes apparent that there’s an impact, be proactive and forthcoming.
Estimation Isn’t the Enemy
If you get nothing else from this article, I hope you at least get a renewed sense of appreciation for the importance of estimation (and a hunger for sushi). Don’t fall into the trap that the agile community frequently sets in how it diminishes the important role they play in strategic planning. It’s one thing to delivery early and often, which should always be the goal of every agile team, but it’s an entirely different thing to be able to predictably put together a strategic growth plan for the business that requires a level of understanding of upcoming delivery commitments , especially for businesses with longer sales cycles.
Strive to become the best estimators possible, using data-driven approaches, and gain your leadership team’s trust by demonstrating proactive communication and an astute eye for early shifts in commitments.
I think I’m in the mood for sushi again. Maybe I’ll try the new restaurant further up the road.