Skip to main content

Organizing teams for success: Horizontal or Vertical?

How do you organize teams for success?  Who do you put together with whom, and who will do their performance reviews?  I don't have answers, but I wanted to lay out the concepts more clearly as food for though...

A recent Java Posse podcast got me thinking about this.  Is it better to organize teams horizontally (backend, middle tier and front end) or vertically (one team creates a product or feature through all three tiers).


A horizontal team is organized where the team is focused on a layer that cuts across more than one application, like this.


A vertical team is organized around getting their product or project done, and cuts across all layers of an application.

My experience in the past few years led me to believe that vertical is always better, but now I think it isn't quite that simple.  Consider this table:

  • Products get done faster because teams don't wait on each other
  • User experience is consistent through a product
  • Core services cost more (reinventing the wheel)
  • Inconsistent user experience between products
  • Services / products are more stable due to a consistent architecture
  • Performance is often better because team members build deep expertise
  • Products take longer to build (teams often wait on each other)
  • Individual team members can stagnate (work is done in silos)

You might want to call bullshit on some of my bullets.  Your list will certainly differ (and I'd really like to know what it is, so please leave a comment.)  You might be part of an organization that fails to reap the benefits while experiencing all of the negatives of both models.  Many do.

Each choice has it's own set of costs and trade-offs.  There are a multitude of "hybrid" solutions that can eliminate some of the costs and increase some of the benefits.

Another Type of Horizontal / Vertical

Reporting relationships affect how we get things done as well.  You can have every member of a team report to one person regardless of their specialization, split them by their specialties, or do a mix.

For example, I work in an organization where all of the QA people report to one director, all of the business analysts to another, and all of the software engineers to a third, with some managers in between for the larger groups.  There are other roles like content production and product development who have their own reporting structures.  That means members of a project team might report to 4 or 5 different people.

Our structure supports consistent practices by role, but lacks focus on individual products.   Individuals don't report to someone who is accountable for delivering products or projects. We have to make up for that by assigning people to projects with responsibility for delivering them.

There are plenty of other ways of going about it.  You could have people report to a manager who is responsible for a product.  That has the advantage of stronger accountability baked-in to the reporting structure.  It has a few drawbacks, such as making it much harder to move between teams and an "us vs them" attitude becoming prevalent.

My conclusion is that there is no one "best" team or organizational structure.  You really need to try more than one to learn what works, and be willing to change them over time.  This can be a very expensive proposition.  Particularly if your employees expect to have one "perfect" structure for all time.  In that environment (which I think is most office environments), every re-org feels like a failure on someone's part.

Refactoring Organizations

If you work in an agile development shop, you understand that it's good to have an inspect-and-adapt loop without your team.  What if we applied that to the organization as a whole, though?  What would "fail fast" look like if we wanted to do it with team and management structures?

I don't know yet, but I believe there are some preconditions.  The main one is to ensure that you can keep a fairly consistent culture in spite of how teams are organized.  People - software developers especially, need some sort of bedrock reality for their day-to-day work if you want them to accept change without losing productivity.

A company to watch in this space is Spotify.  They have deliberately addressed culture and the need for communication through informal networks with their squad / chapter / guild / tribe concepts.  It is a supporting structure for pursuing excellence of craft in software, which lives inside an organization focused on getting product out the door.

We've kicked their concepts around for a while at my workplace, but I can't help feeling that we're missing some key elements.  Guess I'd better read that slide deck again.  Maybe you should too.

Post a Comment

Popular posts from this blog

Agile Performance Management: Why Performance Reviews Suck

Many thanks to Mary Poppendieck, who wrote about this topic in 2004, and proposed a comprehensive solution.  She is the inspiration for much of my thinking on this subject.  She is also a better writer than I am a cartoonist.

Performance reviews suck.  I don't know of anyone who goes into their appraisal without some trepidation.  Your boss is guaranteed spring some surprise criticism on you that is ill-informed or misses the point as you see it.  It's a real challenge not to get defensive about that.

The only thing that makes your own performance review suck less is having to give them.  As a manager, I have dished out quite a few, and some of them went pretty badly.  (To the people at my first management job: Thanks for helping me learn how to get better at them.  Your sacrifice was not in vain.)  Since then, receiving one isn't nearly as gut-wrenching, if only because I try to make it easier for the guy on the other side of the desk.  I've been there, and I know how …

Do. Not. Optimize.

You've probably heard this quote before:
Premature optimization is the root of all evil.
 - Tony Hoare
Speculative optimization is always wasted time.  In the absence of an actual performance problem, you're just burning time that could be better spent on refactoring your code to make it clearer.  This is exacerbated because performance-optimized code is usually harder to read than code which hasn't received such treatment.

Here is what you're doing when you optimize:
Adding code that now must be maintained.Obfuscating the existing code.Spending time writing code that doesn't add value. But what's that you say?  You have the experience and know-how to decide when optimization is needed?  Maybe, but probably not.   The people at Sun and Oracle may or may not be smarter than  you or me, but they certainly know more about optimizing Java bytecode than we do.

For example, some people think that having a large number of classes is slower than the alternative.  This …

Showing Off: How to Do a User Demo

Rather than repeating what has been said elsewhere, here is a nice short post on agile-for-all that covers the basics.

Here are a few things for my own future reference and teams that I'm working with...

Try to keep each demo to 5 minutes or less.   If it's longer than that, it's possible that you should be demoing more than one story.  More likely, you're just being too wordy.

TALK LOUDLY.   No, louder than that.  Louder.  Do you feel like you're yelling?  OK, that's about right.  You need to put your voice in public-address mode for 5 minutes.

Focus on why your audience should care about the story  This is particularly important for back-end work.  For example: Your story generates a feed of XML that will be consumed by another application. Show the output, and point to a couple of salient features in it.  Then be done.

The important part of the above is "show the output."  Showing the end users how to interact with your service is a separate sit-d…