Jun 17

To 10x Your Profits Start With Retrospectives

20% is boring

As many of you know, there’s a technique called a “retrospective” in scrum, or you might call it a debrief. After each sprint the team spends a short amount of time reviewing what went well, what could be improved, and what they should stop doing. The goal is for the team to learn quickly from each sprint.

So, let’s say I really wanted my team to start doing retrospectives, and the team is unwilling. So I share this little bit of research with them:

In a meta-analysis of 46 studies on debriefs (also called after action reviews or lessons learned), Scott Tannenbaum and Christopher Cerasoli found that when appropriately conducted, debriefs can lead to a 20-25% average improvement in performance.  The authors found performance improvements in the use of debriefs with both teams and individuals.  (Thanks to the Center For Evidence Based Management for sharing this study.)

Obviously, based on this data retrospectives and debriefs should be a no-brainer. They are easy and they make your team 20% more effective. But this data still doesn’t motivate most teams to act. Why not?

Because a 20% improvement isn’t enough to get the team to change its behavior.

I need a 10x benefit

This is a general problem for product managers. 20% improvements aren’t interesting to people, whether they’re your customers or your team members. One of my product management rules of thumb is the “Factor of 10 rule” – people aren’t willing to make a change – to a new product, to a new process – unless they get a 10x benefit on a metric they care about.

Back to my problem. I still think retrospectives are a good idea, because I want my team to learn faster. What if I can figure out how to pitch retrospectives as making an order of magnitude difference to something important?

When does 20% equal 10x?

Somehow, I have to make 20% appear to be 1000%.

And that’s what this post is about. (It’s not really about retrospectives.)

Let me tell you one of my favorite stories. At NetIQ we improved the uptime of Windows servers by 20%. That’s nice, but as I said above, it’s not compelling.

But by improving uptime, we also reduced downtime. If you do the math, we reduced it by a factor of nearly 100 – two orders of magnitude! And I can assure you that our customers cared a lot about having less downtime. That’s a story you can tell, and indeed at NetIQ we told that story well.

Can we apply the same kind of thinking to “20% performance improvement?” Somehow turn it into an order of magnitude improvement?

Well, here’s one way. It’s similar to the “downtime/uptime” calculation we did at NetIQ.

Teams improve by learning faster

How fast will the team learn if they don’t do retrospectives? I assume the team’s performance will increase over time naturally, but kind of slowly. Let’s call it 2%. So, if retrospectives increase that rate to 20% … I get a factor of 10x right there.

With this data I can say, “I recommend the team start doing retrospectives – your performance will improve 10x faster.” Retrospectives accelerate the team’s learning, and they accelerate it by a factor of 10. That’s a much better pitch. And learning faster is a benefit that most teams consider important.

How does 10x faster learning affect the bottom line?

But now let’s make the problem a little harder. I don’t just want the team to do retrospectives, which are free. I want the company to buy the team a tool to support retrospectives. And unfortunately, the execs don’t care that much about team learning, no matter what they say about “the company’s people are our most important asset!”

So I have to go to the next step. I need to show a dollars-and-cents business benefit of “improving performance 10x faster.”

What happens when you accelerate your learning by a factor of 10? Well, the whole point of having a higher performing team is to release higher quality products faster.

And what happens when you introduce higher quality products to market faster? You get more revenue faster.

What happens when you get more revenue faster, but with the same team and the same level of effort? All the additional revenue goes straight to the bottom line. So your profit grows, and it grows faster than it would have.

Triple profit for almost no cost

Let’s put some numbers around that. Let’s say that improving the team’s performance results in a 10% improvement in time to market, which results in a 10% improvement in being able to close sales. In any period I get 10% more sales than I would have.

Let’s also assume that our profit margin was 5% before. If I had $10 million in revenue in a quarter, our quarterly profit was $500,000.

With my improved team performance, all the costs are the same, except I’m spending a little bit more on a tool – say that’s $3,000 per quarter. Otherwise, all I’m doing differently is having another short meeting every few weeks. I didn’t add anyone to the team. But now my revenue is $11 million in the quarter because I got better product to market faster. My profit is now $1,497,000.

Improving the team’s performance by 20% resulted in tripled profits!

Oh wait – did I say “triple?” How about 10x my profits?

This is even more impressive if the company is barely profitable. If our profit margin is only 1% before improving the team, then improving the team can increase my profits by a factor of 10.

Unwinding the example

If I want to pitch retrospectives. I’m going to use some of these ideas. If we do retrospectives or debriefs:

  • The team will learn 10x faster
  • Therefore, we will get better quality products to market faster
  • The result will be as much as 10x the profits

I illustrated this in terms of retrospectives. But the same kind of thinking can apply to many other kinds of improvement.

How to turn 20% into 1000%

  1. Find the baseline, call it X. In the example the baseline is the rate at which the team is improving without doing retrospectives. They’ll naturally improve some, but not as much. Let’s say it’s 2% (per year).
  2. Find the improvement to the baseline, call it Y. For retrospectives, it’s 20% per year.
  3. What’s the ratio between the improvement and the baseline (X/Y)? For retrospectives it’s 10. This means that in the retrospective case, the team is getting better 10x faster.
  4. Optional next step: What does that 10x improvement mean in terms of business results? There are a few big ways to improve the business – get to market faster, win more of the deals you get into, get into more deals, open a new segment, increase the total addressable market (TAM) of the segment you’re in. Improving the effectiveness of the development team can impact four of those five.
  5. Figure out how impacting one of more of those factors affects revenues and profits. Especially for profits, and especially if you are starting at a point of low profits, you can often get amazing multipliers for just improving your processes a little bit.


Jun 17

How To Talk To Your Executives About Agile

People have asked me “How can I get my execs to support an agile transformation? They don’t like that they can’t predict anything.”

Reframe the Agile conversation around value

Graffiti on a wall spelling out the word "Value"

Orient the conversation around value, not around methodologies. (Picture credit below)

This is never going to be an easy conversation, but I recommend reframing it from the outset. Don’t talk about scrum or methodologies or the Agile Manifesto. Instead, talk about delivering value to the market and how your products and your company can be more successful.

Among the things keeping executives up at night are the following three questions:

  1. How can I get higher quality products to market faster?
  2. How can I motivate my teams and help them improve?
  3. How can I stop failing in my predictions of what we’re going to deliver?

Help your executives sleep better

If you take these questions to heart, you’ll find there are solutions. And as you implement these solutions, you end up doing agile or something that looks very much like it. (Another way to say this is that you implement agile to get good answers to those questions.)

  • To get higher quality products to market faster, you need to focus on and prioritize adding the highest value capabilities to the products. And you must stop adding low value capabilities. You get two benefits from this. First, because you aren’t working on lower value stuff, the time the team spends is intrinsically higher value. And second, your team is naturally going to be more motivated because they are working on higher value stuff. Focusing on value helps with the first question and part of the second question.
  • I’ve already talked about how to motivate your team more effectively. How do you help them get better? How do you accelerate your team’s learning cycle? You make the cycle of review and learning a lot faster. The “retrospective” is a key part of the scrum methodology for just this reason. The team does its learning as closely as possible to the action. It also helps if you can have the team accomplish things on that short cycle as well. You may need to do clever partitioning of those important things you are working on.
  • There’s only one way to stop failing in your predictions. You must stop trying to predict the future. We all know it doesn’t work. Even if you have a concrete plan, with “good” estimates, you will not hit it. (It’s never been done.) Instead of predicting what you will deliver when, use a value-based approach to talk about what’s coming. You can confidently say “Every release will have the most valuable capabilities we could deliver in the time frame.” You may not know exactly which capabilities those will be in advance. But you can be confident they are the most valuable things you could possibly deliver. So rather than focusing on hitting a schedule – which is an internal target that customers mostly don’t care about at all – you are focusing on delivering value to customers – which customers do care about, which is much more valuable.

Value and velocity – related but not the same

I’ve separated maximizing the value of the results the team delivers from continuously improving the team’s ability to deliver.

These two ideas are often munged together in agile discussions. Indeed, by focusing your team on delivering value fast, the team naturally gets more motivated and faster.
But the second half, of accelerating the learning of the team, is not necessarily about “agile.” The practice of retrospectives or debriefs gives you the other dimension of team improvement – continuous learning. To accelerate continuous learning, you need to make the pace of the retrospectives faster. It’s much better to do learning near the actual experience. And since agile methodologies have short sprints, doing a retrospective for each sprint is a natural way to accelerate learning.

From Value-focused to Agile

If you combine all these thoughts, there are four aspects that combine to make more productive teams:

  • Making sure you’re always working on the most important thing or things
  • Continuous learning
  • Not predicting the future
  • Fast paced iterations

And, if you have those four things, you get four side effects, all very valuable:

  • Agile as a side effect
  • More effective teams as a side effect
  • More revenue faster as a side effect
  • Higher value products to market faster as a side effect

And there’s one other side effect. If you are working in order of importance, you can also push that practice back to the requirements and planning stages. The product managers only need to write requirements and do planning on the next most important capabilities. Requirements for less important capabilities can be delayed until they are needed. This means the development process can start sooner. And that accelerates your time to market even more.

How you can put this into practice today

Here are three things you can do today to start putting these ideas into practice.

  1. Start thinking about value value value, all the time. Only work on the most valuable things you can. Or, to put that another way, don’t work on things that aren’t valuable, even if they are cool or fun.
  2. Make sure your development team understands the value of what they’re working on. You can use the template I mention in this article to help: This New Template Helps You Write Better Product Requirements.
  3. When you talk to your execs about agile, don’t talk about agile, but about delivering value to market faster. Which results in higher revenue and more profits. Execs are typically much more interested in higher revenues and profits than they are in development and project management methodologies.

Image credit

  • “Value” by J.Lightning. Copyright (c) 2011 J. Lightning, Attribution-ShareAlike 2.0 Generic (CC BY-SA 2.0) 

Jun 13

Complexity Bites – Why There Are So Few Good Product Management Tools

I’ve been thinking lately about the characteristics of product management as compared to some other business processes. Some of this is driven by the fact that we have such a lack of tools to help us do our job, while our colleagues have a surfeit of tools – the people running the business have SAP, the sales people have Salesforce.com, the marketers have Marketo or Exact Target, and the developers have a ton of tools for editing, building, tracking bugs, etc. But we PMs are not well served by tools, and in fact live most of our lives in either Excel and Word, or in Jira, or in a combination of those. And none of us are happy with this situation.

No Tools! (photo by flickrminimum, CC2.0 licensed)

Why are so few PM tools available? And why have none of those that exist caught the imagination of product managers to the degree that we have put our credibility on the line and pushed for their installation at our companies?

There are a few reasons, one of which I want to focus on now:

  • PMs don’t have a tool budget. Unlike developers, or the backoffice, there’s no budget for PM tools. But of course, those other disciplines didn’t have a tool budget until there were tools to buy, so that’s a little bit of a post hoc, ergo propter hoc argument (putting the cart before the horse).
  • Maybe product managers are happy with “making-do” with MS Office and wikis and Jira. Well, we’re not that happy, but we manage to live with it. We’re very adaptable! But this again is not an explanation – those tools are difficult to work with for our jobs, and they have lots of downsides that we can all recite.
  • Well, perhaps product management is qualitatively different from the disciplines that have been automated. In particular, it is complex, rather than simply complicated.

Obviously, I am going to riff on the last point – product management is complex. I am using the term complex in a technical way to differentiate it from complicated. Complicated, in this usage, means that there are a lot of moving parts, but they interact in predictable ways. The back office (automated by SAP) is complicated – there is a right answer for how much inventory you have, and you can get to it by following a set of steps. Complexity is a step beyond that – there are a lot of moving parts, but the interdependencies between them, as well as the outside world, result in emergent behavior that can’t be predicted.

Product management is qualitatively different from the disciplines that have been automated

This aligns with my (our?) experience as product managers that we often don’t know what we want until we see it, that no matter how well we conceptualize or prototype a feature or capability, we don’t really start to understand until we see it working, and even then, when we show it to a customer we learn more. And that this happens regardless of what methodology we use to discover, elaborate, document, prototype, design, implement, and test that feature. It’s always possible to have a worse methodology, but it’s never possible to have a methodology that “gets it right the first time.”

These are all characteristics of complex knowledge domains. It’s one reason the Lean Startup approach is so appealing to software product managers – it explicitly says that at the outset you don’t know anything. No matter how good your insights, you have to test them to validate them, and then you’re likely to need to make a pivot as you learn new information.

What does this mean for tools? We’ll cover that in the next post, and I’ll start bringing together some of the threads I’ve been discussing for the past few weeks on creativity, blocks, and the general angst of product management.

In the meantime, what do you think about the proposition that product management is qualitatively different from the disciplines that have been successfully automated, and what do you think are the implications for tools for product management?

For further reading, to whet your appetite and make your head hurt, I suggest looking into Cynefin, a framework for understanding knowledge developed by Dave Snowden. A series of seven blog posts on the origins of Cynefin is collected into this PDF – interesting although slightly difficult reading. (Hat tip to Jean Baraka of Rally Software who turned me on to Cynefin at Agile 2011.)

May 13

Accelerating Change In Your Face

Exponential Change: A Handy Mnemonic

We all know change is accelerating, but, as Peter Diamandis points out, things can be deceptively slow before the exponentials kick in. In this video, in less than three minutes Diamandis shares what he calls the “Six D’s Of Exponentials” – things to look out for as the world changes. Nowadays this change usually starts as Digitization, his first D, and it starts out Deceptively slowly, his second D. I’ll let him share the rest of the D’s.

The most significant of the D’s, from the standpoint of product managers, is the fifth one – Demonetization. These transformations continually make goods and services cheaper, particularly if they can be delivered or implemented digitally. This suggests, at a minimum, that our markets have shorter and shorter lifespans.

Agile Everywhere

On the topic of change, I ran across this second video from Bruce Feiler yesterday, from a TEDx conference last year. It describes how the speaker took the concepts of agile development and applied them to managing his family – chores, bickering, choosing vacation spots – to great success. This is a different kind of change, a change of thinking, that’s the type of thing we’re going to need more of in our future. Many of our beliefs and common knowledge about the world turns out to be outmoded and old-fashioned, ready to be replaced by new ways of thinking. Whether it comes to family dynamics, or teaching, or government, or the economy itself, old ideas are rapidly running out of steam, and failing to produce as much value – both societal and economic – as we need.

Have you run across new ideas or examples of accelerating change in unusual places?

May 13

The Challenge of Roadmaps

Rich Mironov reposted his great article, Magical Thinking and the Zero-Sum Roadmap, about the physics of roadmaps yesterday at the same time I posted my closely-related tidbit about delivering late being better than delivering on time but with low quality. Not a surprising coincidence, since getting product out there is something that occupies our minds tremendously. Rich summarizes his key point as Mironov’s Roadmap Theorem #1: “You can’t put something new into the current development plan without taking out something of equal or larger size.” And that’s a law of physics. Just because the software you write doesn’t need to obey the laws of physics doesn’t mean that the laws of physics don’t apply to its creation.

Predicting the Future – Another Name For a Roadmap

We all love roadmaps, at least our bosses do. They want to know what’s going to happen in the future, and usually they also want to know how much of you’re going to sell, and how each individual feature is going to pay for itself. If you think back to my “Radically Different Manifesto For Agile” post of two weeks ago, you’ll recall my first thesis was “It’s fundamentally better to not try to predict the future. Do your best to predict only as little of the future as you can get away with.” The other points in the post align with that, all based on the ultimate point – predicting the future in detail is impossible. This is a statement we all agree with, right?

Of course, you can predict some big things, what Daniel Burrus in Flash Foresight calls “hard trends.” Computing power will basically double every 18 months – this is one of the most consistent trends in our modern experience, and may even be a law of physics. But, you can’t predict the specifics of those trends – what our computers will look like in five years, for example, or which technology will win the next computing paradigm in 10-20 years.

The Creative Act

So, we know that the details of the future are impossible to predict. But that’s compounded, at least in the software world, by the fact that software doesn’t obey the laws of physics. Another way to say this, as Kurt Leafstrand did in his Medium article Don’t Build, Compose, is that in a software startup, “You’re writing a novel, not constructing a bridge.” Or, not to put too fine a point on it, creating software is not engineering in the same way building a bridge is.

Software is a creative endeavor, more like writing a novel. I talked about this last week in Product Management and Fear. There is no right way to get to the goal, for every new feature in your product you’re starting with a blank page, and you’re likely to go down a lot of rat holes before getting to a form that’s delightful and delivers value. In the article Leafstrand says engineers often need to “tweak the storyline a bit.” I’d go farther – “tweaking” doesn’t quite capture it – I’ve heard authors describe needing five completely different drafts of a novel – different tense, different voice, different characters – before they find an approach that works. As product people, our fantasy is that we’re only a few tweaks away from success, but sometimes we need to shift gears completely in order to make progress. (Hell, every paragraph of this blog post has gone through at least three edits!)

We are therefore put into a difficult place – we want the ability to predict the future (that is, a roadmap), we have constant pressure to add to the roadmap without changing the dates, and we’re writing a novel, in practical terms.

Roadmaps Become Perceived As Reality

And then, unfortunately, the roadmap usually transitions, in peoples’ heads, into reality. Our best guess at predicting the future, which was at best always optimistic, becomes the expectation outside of the product team. In the meantime, inside the product team reality is asserting itself and we are doing our best to deliver the most value we can using agile to prioritize the most important features first, and striving to avoid what I wrote about yesterday (delivering something “on time” but crappy) and instead deliver something incredibly useful, engaging, and that people will buy. On whatever schedule that takes (within reason).

There, in a nutshell, is why product management is not just a fun job, but a tough one. If you can dance that dance, and not get fired, and still get software out, and love doing it, you can be a product manager.


May 13

A Radically Different Manifesto For Agile and Lean

I do not like the agile manifesto. It does not resonate with me as a product manager – someone responsible for guiding successful products to market. It’s very focused on the needs and desires of developers, and much less so on my needs and desires, or more importantly, on those of my customers and prospects. But, I love agile, and I believe agile – and its follow-ons, like lean and kanban – are critical to enable people like me to create and deliver new products that create value in the world.

My Six (Not 99) Theses

What is it about agile that turns me on, while the agile manifesto itself leaves me cold? I’ve made a stab below (and made other points elsewhere) at a high-level explanation for why I like agile and why I think it works. It’s based on some tenets or axioms that lead to the conclusion that an approach like agile is generally superior for delivering value than more traditional approaches like waterfall. I guess these are my version of a manifesto, one that leads to agile and lean when they are applied to product development:

  1. It’s fundamentally better to not try to predict the future. Do your best to predict only as little of the future as you can get away with.
  2. It’s fundamentally better to be responsive to changes, especially as change happens so fast.
  3. It’s much easier to predict how long it will take me to do one thing than it is to predict how long it will take to predict multiple things.
  4. Don’t try to predict where you’ll be a long time in the future. Instead, keep your prediction interval as short as possible – you’re much more likely to be correct. (E.g., if I travel at 50 mph on this road for 15 minutes I can predict with great certainty where I’ll be, but if I have to predict where I’ll be in two weeks’ time, I could literally be anywhere!)
  5. Predicting what the market will want far in the future is extremely difficult, so instead do your best to only have to predict what the market will want in the near future.
  6. “Work in progress” is fundamentally a type of waste – this is a tenet from lean manufacturing and the Toyota Product System, but it also applies to software. Software that’s been developed but is not in use by customers – the definition of “work in progress” for software – is wasteful. Even if it eventually provides value for customers, it’s not providing value right now, leading to an overall loss of total value, which is undesirable. Note: This also applies to requirements – a fully fleshed out requirement that no one is working on is waste, so my recommendation is “don’t do that.”

If you combine those tenets “agile” naturally pops out: Don’t focus on optimizing a large set of work, instead focus on completing the most important piece of that work, as quickly as you can, and deliver it. Since you’re only delivering a small chunk, you don’t have to predict very far into the future. And you’ve chosen the most valuable thing to work on, so you know your market will want it. Because you are working in small chunks, if the market changes its mind, you can quickly change your plans in response. And by always having only a small amount of work in progress, you reduce waste and create more value.

Agile, Upon Analysis, Turns Into Lean and Kanban

In this analysis, though, it’s not really agile that pops up, it’s lean, or more accurately, “Kanban” (at least as practiced in software development). No matter how long your sprints are, some of your features will take longer than the sprint length to deliver to the 80% level. This is a big problem for agile methodologies, because a feature that’s only at the 60% level (which perhaps you can get done in one sprint) doesn’t provide significant business value, and usually neither you nor your customer gains by delivering it at that point. (There’s clearly another blog post in this topic of 60% value versus 80% value.) And therefore Kanban’s approach, which I will simplify as “take your three most important features and work on each one feature until it’s deliverable, then deliver it, and then start on another feature,” does a better job of delivering value to the market and reducing work in progress.

I’ve used the term “lean” above, and I’m really talking about the concepts that come from lean manufacturing. But these ideas are also intimately tied to the “lean startup” mindset as well. One of the key insights of the lean startup is that instead of guessing or predicting the future (all the things I’ve said you don’t want to do), you create hypotheses and test them, getting real data about how customers respond to your product and its features. Agile and lean and kanban again perfectly align with that approach:

  • Get “finished” work out to the customer as soon as possible to find out if they like it – after all, there was a hypothesis that drove the feature or whatever it is, and you want to test whether the hypothesis was correct or not. If so, yay!, and your customers are going to be more successful faster. If NOT, then yay! You have learned something that you can immediately address without letting the issue wither on the vine.
  • Instrument the product so that you can understand how it’s used over its life. This way you can find out what users do with the product, how they’re using it, where they’re running into problems (using the same techniques as webmasters use to assess the usage and obstacles of their websites).
  • Getting features out earlier means that you can get paid for those features faster, whether this is due to existing customers expanding their installations, or renewing their maintenance, or due to new customers coming on board because they heard from their friends (or via an evaluation) that the product does more of what they need better. In other words, more features earlier means a better top line. And the top line is the one that grows the economy – both yours and ours.

Do We Need A Product Managers’ Agile Manifesto?

I don’t know why I feel the need to reinvent things, like the agile manifesto, but there you go. I’d love to start a conversation about this, and if other product managers resonate with my ideas here, perhaps we can refine them and put out our own manifesto on how to deliver more value to market, faster.

May 13

There Is No “Product Owner” For A Commercial Software Product

The Eternal Agile Question: Product Owner or Product Manager?

I’ve been thinking agile lately, and that whole question about the “product owner” versus the “product manager.” Which should you have (or should you have both?) and what do they do? As I’ve mentioned before, the key analytical point to remember is that agile – and the “product owner” concept – came from internal IT development projects, where the project was generally for a particular business unit. You could get a representative of the business unit – a real product owner – to come and represent its needs in a reasonable way. That person’s idea of what was important was considered trustworthy, and so that person could effectively make the prioritization calls. In other words, it was one person, and they knew what the priorities should be, because they were experts in that actual application.

The Types of Projects

Contrast this with two other types of development projects that are out there in the world. There’s the “F-22 Strike Fighter” type of project. In this project, the customer (the U.S. Defense Department, for example) develops a set of requirements that is then farmed out to contractors to implement. The goal of this project is to end up with a product that precisely delivers on the requirements as spelled out in the requirements document, no more, no less, and to do so at high quality and in a reasonable amount of time. And if there is a change to the requirements, it requires renegotiating the contract as a whole. Prioritization in this case is pretty easy, at least in the base case – the contractor has to implement everything in the requirements. That’s it.

Then there’s the type of project that I work on, which is a commercial software product. In this case, the company producing the software, and in particular, the product manager, creates the requirements. (If we’re doing “agile,” of course, we focus on the most important ones initially.) It’s pretty easy to define a whole universe of features that could go into the product, only a fraction of which we are capable of implementing in a timely manner. So the fundamental activity of the product manager, after coming up with this huge list of features, is to prioritize the list with the goal of having the most successful product when it gets into the market. There are lots of tools for this – lean startup is one great approach – but at bottom it involves developing a deep understanding, or at least a deep set of analytical tools, as to what the market really wants, what it is ready to actually buy, and so on.

Product Owner Is The Wrong Abstraction for A Commercial Product

The key point, when it comes to talking about a “product owner,” is that in the third case, commercial products, there is no product owner. There is no one person who knows what the product needs, whose decision is always assumed correct, and who can truly represent the universe of users. The product manager, rather, becomes a channel to the market, the set of users, and has to use techniques and skills to translate what the market wants to what the product should do.

In the spirit of agile, these are admittedly somewhat rough distinctions, but they are qualitatively on the mark.

  • An F-22 type project has relatively little prioritization needs, but it has a lot of need for traceability, especially to testing and quality issues.
  • An internal IT project has more need for prioritization, but there’s assumed to be one person who is capable of doing that prioritization and whose answers to prioritization questions are directly inspired by the usage of the product.
  • Then there are commercial products, where prioritization is a fundamental, deep, and very difficult issue, because there are typically lots of unknowns, there is competitive pressure, and so on. No one person can play a “product owner” role out of his or her head or based on his or her experience, but instead needs to actively engage the market and channel its needs and desires.

Agile Is Powerful, But “Product Owner” – Meh!

I don’t mean to imply that agile is meaningless in the context of a commercial product. I’m only addressing the product owner concept. As I’ve written about several times, an agile process is an intrinsically better approach to developing something of value than a waterfall process, even large features, because its nature is to ensure that delivery of value is front-loaded.

I’m curious whether others have struggled with this product owner vs product manager question, and what you’ve concluded. Please share your thoughts and let’s have a conversation about it.

May 13

Agile vs. The “Big Dog” Feature

(First published back in 2010, this article is still apropos and I thought it was time to revive it.)

Managing big features using agile

This post originated as a comment on the Cranky Product Manager’s blog, responding to her post on agile methodologies. She said

Yes, Agile can speed up the development and improve the quality of small features.  But it’s too often at the expense of the Big Important Work — the heavy lifting, multi-month market analysis and architectural work that lead to REAL customer value and REAL competitive differentiation.

I magnanimously offered my perspective on agile, boiling it down to the key points of:

  • Do the most important things first
  • Be prepared to reprioritize on a regular basis as the environment changes

Or, “spend your resources on the 20% of capabilities that will get you the best return.”

So far, so good. Can’t argue with that as an approach, not just for developing software, but for living life itself. And for the self-help industry, which generates tens of thousands of pages on the Pareto Principle every year. And I’m sure CPM, as we call her, was super-happy that I clarified that.

How waterfall falls down

In contrast, the old way (aka “waterfall”) is more like:

  1. Figure out what you want to accomplish
  2. Determine the most efficient way to accomplish it

It’s easy to see waterfall’s problems when characterized this way:

  • You have to know up-front what your end game is – which makes it hard to respond to market changes
  • The most efficient way of building the full product is not necessarily the one that front-loads the value, so often low-value items are completed and high-value items end up deferred
  • You do a lot of work up-front to document things that the developers never get to

Notice that’s not how agile talks about itself. The agile manifesto talks about working code vs. documents, and interactions over tools (it does cover “responding to change”). In fact, agile, as far the methodologies and manifesto go, is solely focused on programming. (This is changing some – I just listened to Kent Beck’s talk at the Ruby On Rails conference last year, and he’s come up with a very different characterization of the goals of agile: it provides accountability, responsibility, and traceability.)

How agile lets you tame the big dog

Now, given my characterization of agile’s – indeed, life’s – key goals, you can then look at agile  methodologies simply as one way to accomplish those goals. But what if, as CPM fears, the most important capability (call it The Big Dog) takes longer to deliver than a sprint or two, and requires visits to lots of customers to understand their problems, and lots of reviews with customers to see if we’re solving their problem?

Clearly, you still have to do the Big Dog. CPM should be able to tell you why. And if the methodology doesn’t give you a way to do it, then the methodology won’t work for that product.

But chances are the 80/20 rule applies to the Big Dog, just as it applies to everything else. And this large monolithic capability can be broken down sensibly into multiple passes through the “smallest thing that could possibly work” approach. Does this require the PM to keep ahead of the development organization? Yes. Is that any different from the old days? Yes and no. The PM needs to figure out the most important part of the Big Dog (the 20%), and make sure it’s understood, there are good user stories, it’s designed, architected, etc., extremely well. After all, that’s where most of the value is going to come from.

But the PM doesn’t need to document the rest of the 80% until later – if at all. In fact, it’s likely that finishing the 20% of the Big Dog prioritized to the top of the project leaves something else – the Medium Kahuna – as the next important item to accomplish. There may be some additional Big Dog-related capabilities that are “nice to have” – and they’ll be prioritized into the rest of the project, if there’s time after getting the Medium Kahuna delivering its value.

“Now wait,” you (or the CPM) say, “I can’t live with only 20% of the Big Dog – I need 100% of it – or at least 80%” And I say this is where the beauty of the agile mindset comes into play. If you’ve completed 20% of the Big Dog, and have the rest of the Big Dog as well as the Medium Kahuna in your backlog, at this point you can decide which is more important, and decide which one to do. You’re already delivering 80% of the value of the Big Dog – now you can decide if you really need to take that up to 90%, and leave the Medium Kahuna on the table, or vice versa. You have control.

Agile can’t solve every problem, but can tame a lot of big dogs

Agile is not a silver bullet, and it’s hard to get right, but if it helps you focus on putting first things first and executing on the 80/20 rule, it’s done its job.

Your thoughts?