AoAD2 Chapter: Discovery

This is an excerpt from The Art of Agile Development, Second Edition. Visit the Second Edition home page for additional excerpts and more!

This excerpt is copyright 2007, 2021 by James Shore and Shane Warden. Although you are welcome to share this link, do not distribute or republish the content without James Shore’s express written permission.


Optimizing teams make their own product decisions. How do they know what to build?

Whole Team

Partly, they know what to build because they include people with product expertise. Those team members have the background and training to decide what to do.

But the fact is, at least at the beginning of a new product, nobody is 100% sure what to do. Some people pretend to know, but Optimizing teams don’t. Their ideas are, at best, very good guesses about what will lead to success.

The job of the Optimizing team isn’t to know what to build, but to discover what to build.

So the job of the Optimizing team isn’t to know what to build, but to discover what to build. Steve Blank, whose work was the basis for the Lean Startup movement, put it this way:

[T]he task is unambiguous—learn and discover what problems customers have, and whether your product concept solves that problem; understand who will buy it; and use that knowledge to build a sales roadmap so a sales team can sell it to them. And [you] must have the agility to move with sudden and rapid shifts based on what customers have to say and the clout to reconfigure [your team] when customer feedback requires it. [Blank2020a] (app. A)

Steve Blank, The Four Steps to the Epiphany

Steve Blank was talking about startups, but this quote applies equally well to Optimizing teams. Even if you aren’t selling your software! No matter who your customers and users are—even if they’re Keven and Kyla, who sit in the next cubicle over—your job is to figure out how to bring them value. And, just as importantly, how to do so in a way they will actually buy or use.

Validated Learning

I can’t count the number of times I’ve had a good idea, put it in front of real customers or users, and found out that it didn’t work out. Sure, they would tell me they loved the idea when I told them about it. Sometimes, even after they tried a prototype! It was only when I asked people to make a real expenditure—of time, money, or political capital—that I learned my “good idea” wasn’t good enough.

Product ideas are like a perpetual motion machine: if you believe hard enough, and have enough inertia, they look like they’ll last forever. Put a real load on them, though, and they grind to a halt.

Blind Spot Discovery
Real Customer Involvement

Validated learning is one of your best tools for testing ideas. I discussed it in the “Validated Learning” section, but to recap, validated learning involves making a hypothesis about your market, building something you can put in front of them, and measuring what happens. Use what you’ve learned to adjust your plans, then repeat. This is often referred to as the Build-Measure-Learn loop.

To truly validate your learning, you need real customers (or users) and real costs. If you show what you’ve built to people who aren’t part of your target market, you’ll get feedback, but it might not be relevant to your actual situation. And if you don’t ask them to commit something in exchange, you’ll learn more about people’s desire to avoid hurting your feelings than about the actual value of your idea. Everybody will praise your idea for a luxury vacation...until you ask them for their down payment.1

1Then it’s all, “Oh, I don’t have time,” “I couldn’t leave my chihuahua Fluffles all alone,” and “I hate tropical sand. It’s rough and irritating, and it gets everywhere.”


Adaptive Planning

Every time you go through the Build-Measure-Learn loop, you’ll learn something new. To take advantage of what you learned, you’ll have to change your plans. As a result, Optimizing teams tend to keep their planning horizons short and their plans adaptable. They keep their valuable increments small so they can change direction without waste.

Valuable increments (see the “Valuable Increments” section) aren’t just about features and capabilities. Remember, there are three common categories of value:

  • Direct value. You’ve built something that provides one of the types of value described in the “What Do Organizations Value” sidebar.

  • Learning value. You’ve built something that helps you understand your market and future prospects better.

  • Option value. You’ve built something that allows you to change direction for less cost.

For Optimizing teams, learning and options are just as important as direct value. In the beginning, they can even be more important than direct value, because they allow the team to avoid wasting time building the wrong things. Every Build-Measure-Learn loop is an example of a “learning value” increment.

Options thinking is also common in Optimizing teams. The future is uncertain, and no plans are set in stone, so Optimizing teams ensure they have the ability to adapt. They do so by thinking about future possibilities and building “option value” increments. A prospective analysis, described in the “Prospective Analysis” section, is one way to start identifing those options.

Options are also an important technique for managing risk. If your prospective analysis shows a substantial risk—for example, a competitor providing a less lucrative, but more attractive pricing model—you could build an option that allowed you to change your pricing model with the flip of a switch.

Another sort of option involves deadlines. Although Optimizing teams avoid arbitrary deadlines, sometimes value depends on releasing before a certain date. For example, video games need to be delivered in time for the holiday season, tax software needs to be updated yearly, and new regulations can have strict deadlines with harsh compliance penalties.

To meet these deadlines, Optimizing teams will often build a “safety” increment before embarking on a more ambitious idea. The “safety” increment fulfills the demands of the deadline, in a minimal way, leaving the team free to work on its more ambitious ideas without worry. If those ideas doesn’t pan out, or can’t be completed in time, the team releases the “safety” increment instead.

For example, reviewer Bill Wake shared the (possibly apocryphal) story of a printer company that needed to deliver a red-eye removal feature for a new photo printer. The hardware had a strict release date, so the software team started with a primitive red-eye algorithm, then worked on a more sophisticated approach.

Experiments and Further Reading

There’s much, much more to deciding product direction than I can cover in this book. Opportunities for further reading abound; look in the product management category. Three places to start are Marty Cagan’s Inspired: How to Create Tech Products Customers Love [Cagan2017]; Luke Hohmann’s Innovation Games: Creating Breakthrough Products through Collaborative Play [Hohmann2006]; and David Bland and Alexander Osterwalder’s Testing Business Ideas [Bland2019].

The point to remember is that, in addition to normal product management, Optimizing teams engage with their customers to understand their market and validate their ideas. They exist to learn as much as they do to build, and the flexibility of their plans reflects that focus. The Lean Startup movement calls this customer discovery and customer validation.

For much more detail about these ideas, see The Startup Owner’s Manual. [Blank2020b] It’s an updated version of Steve Blank’s book, The Four Steps to the Epiphany. [Blank2020a] Blank’s ideas, combined with Extreme Programming, formed the basis of Eric Ries’s Lean Startup movement. [Ries2011]

As you can imagine, The Startup Owner’s Manual is focused on startups, so its advice will need customization to your situation, but Optimizing teams have a lot of similarities to startups. A successful Optimizing team isn’t just carrying on with the status quo. If it were, Focusing and Delivering fluency would be sufficient. Instead, it’s seeking ways to lead its market and develop new markets. Lean Startup ideas, including the foundational ideas of customer discovery and customer validation, are a key part of how you can do so.

Share your thoughts about this excerpt on the AoAD2 mailing list or Discord server. For videos and interviews regarding the book, see the book club archive.

For more excerpts from the book, see the Second Edition home page.

If you liked this entry, check out my best writing and presentations, and consider subscribing to updates by email or RSS.