My take on #NoEstimates (now)
I decided to create this post when I got triggered by the following tweet of Tim Ottinger:

For the sake of the argument (the point I want to make) I will not dive into the question ‘What is estimating?’. Instead I will focus or the working environment Tim describes in which he doesn’t need to estimate and a main problem people have involving estimates.
From Tim’s tweets I conclude that the following is in place:
- Safe working environment
- Predictable flow
- CI/CD
- No shareholders/stakeholders requiring an idea when what will be delivered
- No traditional marketing program where promises are made/expectations are set for the customer.
Seems to me that when no-one needs to know what will be built by when and when it is not required to know where the money is being spent on in a certain quarter or year to get budget, that then there is no reason to estimate.
Having said all of this: above example with a predictable flow would make estimating extremely easy. Estimating and estimates probably never were an issue in above example to begin with. This is not the environment where people struggle with estimating and the expectations around estimates.
As far as I have experienced and witnessed and also as described in Vasco Duarte’s book #NoEstimates, people have issues with estimating when:
- There is an unsafe working environment, for example: estimates and plans are seen as set in stone by management. Uncertainties are ignored, which leads to unrealistic expectations.
- People have trouble with estimating. If this isn’t recognized and no corrective measures are taken, false expectations are created.
Both are deadly for an organization. The way estimates are treated is a symptom of a dysfunctional organization. Taking away these dysfunctions will take away the issues with estimates automatically.
So my point here is: the solution to the issues with estimating is NOT to stop estimating. Even worse: stopping with estimates in such a dysfunctional organization will only do more harm, because it is apparently enforced. #NoEstimates would make things worse.
When people know that their efforts in estimating aren’t translated to commitments, when they know they are in a safe working environment, they will not have issues to give estimates.
I can also see the same in Vasco Duarte’s #NoEstimates book. It says on the cover: “HOW DO YOU STREAMLINE YOUR PROJECT MANAGEMENT SO THAT YOU HIT RELEASE-DATE DEADLINES EVERY TIME”. This all without estimating. It also mentions two items that are not in place in the example throughout the book:
- Marketing sold something without having clear view on magnitude of work and as a result time and cost required.
- People work in an unsafe environment, where estimates based on insufficient data is being transformed into an unrealistic planning, which then turns into a commitments.
The #NoEstimates book comes with a magic pill. The PM convinces the customer that the project should be de-scoped. Essentially the customer is told that during sales the company over-promised and that new insights tell them that they can’t deliver everything in time. And the customer accepts this notion. This has nothing to do with making deadlines all the time. I can meet every deadline whenever I have a customer willing to see scope reduced!
However, decreasing scope comes with consequences:
- A customer doesn’t get everything she wants and pays for or
- A customer gets the complete functionality at a later stage, probably also with higher costs as more effort is required to build it than initially planned.
Not my way to always be on time!
On top of that: the project started off as a rigid waterfall project with software to be deployed at the end and gradually turned into a project that delivered in iterations and with increments. Surely a great call, but unrelated to not estimating.
Vasco also comes with certain practices where you can argue if these are really not including estimates, but like I said at the top of the post, I will not dive into that. Point is however that these practices are not solving the issues in the end. The practices that solved the issues were:
- Acceptance of customer that scope is decreased
- Drastic move from rigid waterfall to delivery in iterations and with increments.
Now my bottom line:
- when no-one needs to know what will be built by when and when no-one requires to know where the money is being spent on in a certain quarter or year to get budget, there is no reason to estimate. But it seems to me this is not what is being discussed under #NoEstimates. It is a non-issue to me.
- A main issue within #NoEstimates is the way estimates are misused. This is a symptom of a dysfunctional organization. Taking away these dysfunctions will take away the issues with estimates automatically.
- When you are becoming so predictable that it doesn’t make sense to use story points as estimates anymore, why would you continue doing that? You can just as well estimate and plan based on historical data. So here the topic of story point estimation is addressed, which is basically moving at the fringes of the #NoEstimates debate.
For me #NoEstimates is not about topic 1. When estimates don’t have added value regarding decision making etc: don’t do it, simple. Topic 3 is more interesting, as it looks for alternatives for story point estimating, for example by replacing it with using historical data to estimate. Topic 2 however: I have yet to see a way how this dysfunction can be resolved by removing estimates.
Like I said: this is my evolved view on the # and the reason for me asking questions and having a critical view. But I like to be proven wrong!
Originally published at ageling.wordpress.com on August 25, 2017.
