Black and White stance on #NoEstimates considered harmful

#NoEstimates
Hashing out different opions of NoEstimates

Black and white stance on #NoEstimates considered harmful

In my last post, I shared how Woody Zuill started the #NoEstimates movement with a clever experiment. Furthermore, I shared many taking part in the discussion from both sides agree the dialog is often quite unproductive. Quoting Ron Jeffries "unproductive discussion", "unreasoned-name calling" and "ad-hominem arguments". One part of what fuels the fire is #NoEstimates proponents who take an aggressive black and white stance on the topic, claiming estimates are always bad. Even though Woody, originator of the idea, openly shares that:

In this post, I will cover why taking this stance is harmful, cover what kind of interactions are feeding the fire, and ask what's the purpose of claiming Estimates are always bad.

Why is the black-and-white stance considered harmful

Many who actively take part in the discussions, Ron Jeffries, Your Daily Scrum podcast, Peter Kretzman and others openly agree that keeping the discussions on these negative terms is not helpful. Let's cover why that is and what's the impact.

Let's start by looking up how leadership is defined. On Wikipedia it says that leadership is:

..a practical skill encompassing the ability of an individual, group or organization to "lead," influence or guide other individuals, teams, or entire organizations.

Being a leader means followers support your cause. Followers look to their leader for guidance and take on or repeat their behavior in some ways. That is why people in leadership positions need to acknowledge their role and live up to the responsibility that comes with it.

For this reason, it is:

  • critical leaders on both sides of the discussion take care how they act and what they say

Why, you might ask?

  • Because leaders influence their followers and shape what kind of interactions take place!

So when some of the key leaders of the #NoEstimates movement show the following behavior repeatedly, it fuels the fire:

  1. Claiming "estimates never work", i.e. Faulty generalization fallacy
  2. Using examples of how not to estimate to support claims that "estimates don't work", i.e., making a Straw man fallacy

Why is this harmful?

It is harmful because when leaders show this kind of behavior, what happens is that followers take up the false claims, the aggressive tone, and the attitude their leaders display. This further damages the discussion and divides people. As seen all over online forums where followers make the same false claims, it seems to pick a fight.

Let's take a short example to explain the impact. A few months back, I read a post on Reddit about estimates, with lots of comments, and lively discussions. There were at least 4-5 comments (around 10-15%) from users picking a fight adding comments along the line, you know, 'estimates don't work,' essentially ridiculing the use of estimates. Given the response these comments got, it is clear many have grown tired of these arguments, and so much time is being wasted on unproductive discussions.

So what kind of messages are feeding the fire?

What kind of messages are feeding the fire?

When looking for answers to why so many in the software development community are making this claim, I came across tweets from Vasco Duarte, author of the #NoEstimates book. Vasco released the book in 2016 and is highly active on Twitter under the #NoEstimates hashtag. Nearly every day, Vasco posts multiple tweets under the #NoEstimates hashtag. His tweets are created to grab attention, dismissing the value of estimates and ridiculing people who use them. Seeing the following tweet, I wondered if he thought estimates never work. So I asked him the following question to understand where he stood on the topic:

I thought Vasco's answer below was quite positive because he is accepting it's valuable to have a tool '..allowing teams to make decisions about options that come up during backlog discussions'. In addition, he does not deny estimates provide answers.

To confirm this understanding, I followed up and asked:

His response:

What Vasco is saying is that estimates:

  1. only work for people who use estimates for "..blaming, pressuring others, etc.."
  2. do not work "for containing costs or hitting a deadline"

Quickly browsing his tweets and looking for similar statements, Vasco is saying estimates never work.

That is a bold statement.

Why? Because logically making this claim takes only a single case, one example, where an estimate is used successfully to show statements making this claim are false. A post Debunking the statement' estimates never work' shows these statements are false.

What's next?

In the next post, I will dive deeper into the purpose of repeating these false claims, keeping the polarization in our community rampant. The following post, Did you get caught in the #NoEstimates trap?, shares an inconvenient truth about the methods we see used.

About Smart Guess We help teams educate stakeholders by providing actionable answers to common misconceptions about estimates. So more teams can use estimates to their advantage, and fewer are held accountable for delivering on impossible deadlines.

Recent Posts

Share your thoughts