Should I use hours to estimate my tasks in Scrum?

My new recommended “starter” “complementary practice” for new Scrum teams is to simply create tasks and use the “number of tasks remaining” for a burndown.  I encourage them more to focus on PRI (potentially releasable increments), Sprint Goals, and achieving a moderately consistent level of skill at meeting their Sprint forecasts (used to be called “Sprint commitment,” it’s been re-named in the Scrum Guide).  I also caution heavily against trying to achieve perfect forecast accuracy as that’s a fool’s errand in complex domains.

Using hours for tasks can lead down some really bad roads, most notably:  Former PM’s turned SM’s and other organizational members who try to apply PMI tactics (100% utilization, tracking actuals, etc) tactics to complex software development.  By preferring “sticking to the plan” over “responding to change”, they are completely violating Agile and Scrum.

This same bad road can also lead companies into think that “schedule/scope/cost” is an optimum model for software development.  As far as I’m concerned, schedule/scope/cost is a dead, failed model for software.  

Now, using hours for tasks doesn’t have to lead down those bad roads — but in my experiences, they usually do.  Let’s not forget, Scrum used to require hours for task estimation, many years ago, but the Scrum experiences of the wider community over 20 years has spoken on the topic — hours is not always optimum.  I would go farther than that and say, at the Sprint task level, it’s usually NOT optimum.

Given the above, I’ll leave it as an exercise to others to describe where using task hours might not lead down those bad roads.

_____
Charles Bradley
Professional Scrum Trainer
Scrum Coach-in-Chief
http://ScrumCrazy.com

New Courses from ScrumCrazy.com:

  • Scrum For Executives
  • Agile Requirements: Product Owner and Team Collaboration Techniques
  • Scrum Product Owner: Techniques for Success
  • Evidence Based Management for Software Organizations(TM)
    • Class for Software Development Managers and Executives

If you’re interested in any of our classes, training, or coaching services, feel free to contact us.

Breaking News!! Jeff Sutherland’s Scrum Inc introduces “Scrum At Scale” — see it at Agile 2014 session

I was lucky enough to get a preview of the “Scrum At Scale” approach from Scrum Inc a few days ago. In short, it’s a model for conversations around how to think about scaling Scrum in the enterprise. The model is modular, and it is very clear that this approach is more lightweight and flexible than other Agile scaling approaches that get a lot of attention. Alex Brown of Scrum Inc, the Product Owner for the model, as well as Jeff Sutherland, are very adamant that this this not some cookie cutter recipe or methodology to scale Scrum. It’s different than other approaches, in that it’s a model for conversations around inspecting and adapting toward success with Scrum at Scale.

I’m told that the slides for the presentation at the link above will be posted there within the next couple of days, and possibly sooner. The slides will add a lot of detail that the main graphic doesn’t give. I will also add that there is some nuance and detail not included in the slides either. As such, I recommend try to attend one of their live or recorded video presentations to get some richer nuance…. or….

They are also presenting on this topic at Agile 2014 next week. If you’re going to be at Agile 2014, I highly recommend you put their session on your “must see” list.

The model gives a “big picture” view of Scrum in the enterprise, but it also dovetails nicely with the many years of work that Jeff Sutherland and others have put into their Scrum Patterns efforts. As you may know, I’m also a fan of this Scrum patterns concept, and you can see an example of that work on my website — Daily Scrum Patterns.

It’s worth mentioning that I have no business relationship whatsoever with Scrum Inc, so I’m not in any way incentivized to advocate for their approach.  I’m only endorsing it because I believe in the approach and in it’s future.

I suspect that this work will be a game changer in the Scrum scaling space, which doesn’t surprise me, really. It *is,* after all, coming from a company run by the co-creator of Scrum! Nice work Alex Brown, Jeff Sutherland, and Scrum Inc!

Is SAFe(tm) Bad? Is it unsafe? Is it Agile? Are there alternatives?

For much better alternatives to SAFe(tm), see this page.

All my personal/professional opinion.

I’m not a big fan of SAFe(tm). I haven’t yet had time to sit down and detail all of the problems I have with it, but I’ll hit a couple.

1. It’s not Agile at all. It’s a sales strategy.

My biggest problem with it is that it condones old, out of date, and dysfunctional practices that don’t enhance Agility. It is essentially a hybrid approach of Waterfall and Agile, along with a lot of baggage from RUP. This probably shouldn’t surprise anyone since the creator, Dean Leffingwell, was a big salesman/evangelist for RUP. The biggest baggage from RUP is the complexity of a zillion different roles and the fact that SAFe is a “slice and dice” methodology. It’s not a framework. I think the “slice and dice” thing is really just a slick sales strategy. It allows those selling SAFe to immediately disown any practice of SAFe that a potential client complains isn’t Agile or won’t work. It also means that any tiny subset of SAFe is still considered to be SAFe. As such, I just consider this a sales strategy to sell more billable hours. This strategy was also used in RUP, and yet… over the years… which has prospered more? RUP or Scrum? Scrum is a framework, so there is no slice and dicing of the framework itself.

2. It misleads people into thinking that it uses Scrum at the team level.

It claims to use Scrum at the team level, but then completely sells out Scrum in so many ways. It sells out the Product Owner role by giving control to all manner of people over the Product Backlog contents, something Scrum expressly forbids. It sells out the Scrum Master role by suggesting it’s a 25% time commitment. Then, it completely sells out the Development Team by creating Ivory Tower architects.

3. To date, no Agile Manifesto author has endorsed it. That should tell you something right there.

This one is self explanatory. :-)

The reasons I don’t like it are covered in way more detail in these reviews of SAFe by other people who are sharp enough to tell the differences between SAFe and other approaches, as well as the history behind similar practices and approaches

For much better alternatives to SAFe(tm), see this page.

Great Article from Scrum Co-Creator Ken Schwaber on “Agile Value”

Ken Schwaber’s written a great article on Agile Value .  He talks about how value can be defined in Agile, and the article also covers the difference between traditional project management metrics and Agile project measurements.

On a related note, see Common Project Management Metrics Doom IT Departments to Failure

Scrum and Agile Adoption: Backsliding into Waterfall Habits

It’s very common, when an organization is in the Shu level of learning Agile or Scrum, for it to fall into old, bad, waterfall habits.  Today I’d like to talk about two bad waterfall habits:  tracking so called ‘individual velocity,’ and tracking actual effort expended on a task or story.

First, tracking some sort of metric like “individual velocity” is probably an excellent way to completely sabotage your project/product and it’s also a great way to kill your Agile adoption.  A key concept in the Agile Manifesto principles, as well as in Scrum, is team work and “self-organizing” teams.  Self organization is generally the ability for a team to create and execute their own plan of work(Sprint Backlog), as well as decide “How” to do their work.  Whenever there is a single entity (individual on or off the team, department, etc) who strongly influences or makes unilateral decisions for how a team works, there is, by definition, no self organization.  Tracking individual velocity, or any similar “individual incentive” (this can include raises, performance reviews, awards, etc) does not encourage team work at all.  In his book, Agile Estimating and Planning, Mike Cohn says it this way:

“If I am forced to choose between finishing a story on my own and helping someone else, what incentive does measuring individual velocity give me?  Individuals should be given every incentive possible to work as a team. If the team’s throughput is increased by my helping someone else, that’s what I should do. Team velocity matters; individual velocity doesn’t. It’s not even a metric of passing interest.”

Tracking actual effort expended is another really bad waterfall habit.  Tracking estimates and actuals is just another example of where Common Project Management Metrics Doom IT Departments to Failure.  Mike Cohn, an Agile and Scrum thought leader, and a former project manager himself, says this about tracking estimates and actuals:

“On a project, it is far more useful to know how much remains to be done rather than how much has been done. Further, tracking effort expended and comparing it with estimated effort can lead to ‘evaluation apprehension’ (Sanders 1984). When estimators are apprehensive about providing an estimate, the familiar ‘fight or flight’ instinct kicks in, and estimators rely more on instinct than on analytical thought (Jørgensen 2004).

Tracking effort expended in an effort to improve estimate accuracy is a very fine line. It can work (Lederer and Prasad 1998; Weinberg and Schulman 1974). However, the project manager or whoever is doing the tracking must be very careful to avoid putting significant evaluation pressure on the estimators, as doing so could result in estimates that are worse rather than better.

Additionally, keep in mind that variability is a part of every estimate. No matter how much effort is put into improving estimates, a team will never be able to estimate perfectly. Evidence of this is no further away than your morning commute to work. There is an inherent amount of variability in your commute regardless of how you travel, how far you must go, and where you live. If you drive to work, no amount of driving skill will eliminate this variability.”

In a later post on the ScrumDevelopment Yahoo Group, Mike summarizes it this way:

” I’d say you shouldn’t do it because it doesn’t add value commensurate with its cost. Don’t argue with your bosses that it ‘adds no value’ because
comparing what you originally thought a task would take with what it did take can help make you a better estimator.

But, it can be time-consuming to track actuals, especially for a full team where some on the team are probably already decent estimators.

Because Scrum already has solid mechanisms for monitoring whether all the work gets done in a sprint (high team commitment, daily burndown charts, daily scrum, and so on), Scrum does not have the same reliance on early and accurating estimating that a predictive or waterfall approach does.

So–the cost to gather actuals is the same in Scrum or waterfall. The benefit in Scrum is greatly reduced.”

So, be very careful when backsliding into old waterfall habits.  It usually happens in small doses, which is why many Shu level adoptions don’t notice it, especially if they don’t have a skilled Scrum Coach or highly experienced Scrum Master around.  The other thing to keep in mind is that, even if you do have someone skilled around, this old adage still applies:  “You can lead a horse to water, but you can’t make ‘em drink.”

What kind of backsliding into old waterfall habits have you seen?  What do you suggest be done about them?  Sound off in the comments below!

Great 2 Minute Video on the Product Owner Role in Scrum

SSW of Australia has just put out a great two minute video on the role of the Product Owner in Scrum.

Check it out here(might take a minute to load):  http://tv.ssw.com/3244/what-is-a-product-owner

I’ve added it as #4 under Scrum on the ScrumCrazy.com’s list of preferred Scrum resources.

Great blog post on why the Scaled Agile Framework is “Scaled Amateur Hour” and definitely NOT Agile –

Scrum trainer Dominik Maximini has written an excellent, objective, but critical article on the Scaled Agile Framework.  While my own criticisms would probably be both somewhat similar and different to Dominik’s, he’s done an excellent job of laying out the case against SAFe, and I agree with all of his conclusions.

I highly recommend you give it a read.

Follow

Get every new post delivered to your Inbox.

Join 283 other followers

%d bloggers like this: