Choosing to do a Project: Three Gates

I’ve had this mental model for a while, but it solidified a year ago after explaining it at work, and I wanted to use it again in a Twitter debate shortly afterwards. Twitter’s the wrong place for real explanations though, so here it is.

This model will not solve your disagreements for you, and it’s not a substitute for calculating the NPV of your possible projects. Instead, it’s a quick shorthand that might make discussions more useful, and show you something about your team.

What makes something worth doing? Why would you choose project A over project B? Or why would you choose to make subject X mandatory at school, but not subject Y?

When you have limited resources, and must make difficult, contested choices over what to do, it’s normal for the concerned parties to have different opinions about the right choice. Discussions over every choice can take serious time, so it’s useful to be able to quickly classify your options.

One way to classify them is to ask how many of these sequential gates an option passes through:

Gate 1: Is the result of the choice worth having at all? Are the benefits real?

Gate 2: Do the benefits outweigh the costs? Is there a net benefit to taking this action?

Gate 3: Is the net benefit high enough, given other possible choices?

To clarify these gates, here are a few examples of what does and does not pass the first gates, and ways to think about them.

 

Gate One: Is the End Result Useful?

Starting with some examples that would pass:

  • A slightly better login page, that saves all of your staff two seconds every day. It’s a small benefit, but it’s real, so this passes.
  • Making changes to an internal web site, to make it easier to navigate. This won’t necessarily save much time, but if it saves some time, it passes.

And some examples that wouldn’t:

  • Switching an automated task to start at 3pm instead of 4pm, when it’s not time critical. This doesn’t pass, as it doesn’t produce any real benefits.
  • Starting to gather data on things that don’t matter. If you can’t do anything with the data, or don’t care about it, then it’s a distraction, and not worth having.

Another way to consider whether some project passes this gate is to ask, “If you could have all of the benefits of this project for free, right now, would you want them?”

 

Gate Two: Is there a Net Benefit?

Some things that would probably pass this include:

  • Creating a product, when you already have customers committed to paying for it.
  • An improvement to an internal process that would take 100 hours to complete, but would eliminate the need for one member of staff to work full-time on the process.

Some things that might not pass this gate include:

  • Improving an interface to an internal system, when you plan to replace the whole system within three six months.
  • An improvement to an internal process that would take over a year to complete, but would eliminate the need for one member of staff to work half-time on the process.

Another way to think about this gate is to ask, “If you could somehow pay all of the costs of delivering the project today, and have it go live tomorrow, would you?”

 

Gate Three: Is this One of the Best Options Available?

This is where the portfolio of options available becomes relevant. To pass this gate, a project needs to be not just worthwhile, but more beneficial than the other available options. You can’t work out if a project passes this gate in isolation; it has to be done by comparing it to other projects. (If you have no other projects ready to start or under way, then any project that passes gate two also passes gate three, since it has no competition.)

All other things equal, a project that will save ten hours per week of manual work is not as beneficial as one that will save twenty hours per week. You may want to do them both, but if their costs and risks are the same, you should obviously do the with the biggest benefits first.

When you decide that some potential project does pass gate three, that you want to work on it now, you’re making an implicit assertion of your priorities. As well as the costs and benefits, you’re also happy with the risks and the timescales. Depending on the situation you’re in, you might prefer smaller, quicker, more certain wins over bigger, slower, less certain wins. 

 

The Gates in Use

I found three different uses for this model: firstly, quickly communicating my thoughts about a suggested project; secondly, working out whether my team understood what we were trying to achieve; and thirdly, assessing how close we were to exhausting our possibilities for improvement.

The first use I found for this model, and what cemented it in my mind, was explaining why we wouldn’t be implementing an idea that someone brought to me. I spend a lot of time thinking about how to improve processes and systems, and it’s a habit I appreciate in people who report to me, so I encourage it. Of course, one thing that will quickly put people off taking ideas to their managers is feeling ignored.

So when someone brought me an idea, I would look at which gates it passed, and use that to frame my communication about the idea. If I explained that an idea passed gate two—that I thought it was worth doing—but not gate three, I was acknowledging that the idea was interesting, and potentially worth doing, but I wasn’t committing to doing it at any particular time. This worked much better than, “We’re not doing this because…”.

Secondly, I used this as a way of thinking about my own team. I was never bothered about ideas that didn’t pass gate two or three; it takes time for people to develop good intuitions about costs and benefits. On the other hand, if someone brought me ideas ideas that didn’t pass gate one, so weren’t worth having at all, I had to question whether they understood our business priorities. Why did they think the benefits of the projects they suggested were worth having? How could I get them thinking about our real priorities?

Lastly, the number of potential projects I have that pass gate two, but haven’t been started yet, tells me something about the state of the process or processes I’m responsible for.  If the big wins have been accomplished, and the net benefits of the remaining potential projects are getting smaller, it may be time to consider strategy and resources. Even if there are still benefits to be had from the remaining potential projects, it may still be a useful to change the goals and organisation of the team. Of course this is a big step, and probably can’t be done without wider internal discussions.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.