Product Owners need 4 things


To be an effective Product Owner – and that includes product managers and business analysts who are nominating work for teams to do – you need at least four things. You may well need more than these four but these are common across all teams and domains.

  1. 1. Skills and experience

There is more to being a Product Owner than simply writing user stories and prioritising a backlog. Yes, you need to know how to work with a development team and how to work in an Agile-style process. Yes you need to be able to write user stories and acceptance criteria, perhaps BDD style cucumbers too; yes you need to be able to manage a backlog and prioritises and partake in planning meetings.

But how do you know what should be a priority?
How do you know what will deliver value? And please customers? Satisfy stakeholders?

Importantly Product Owners need to be able to do the work behind the backlog.

Product Owners need to meet people, have the conversations, do the analysis and thinking behind those things. Any idiot can pick random items form a backlog but it takes skills and experience to maximise value.

Product Owners need to be able to identify users, segment customers, interview people, understand their needs and jobs to be done. They need to know when to run experiments and when to turn to research journals and market studies. And that might mean they need data analysis skills too.

If the product is going to sell as a commercial product you will need wider product management skills. While if your product is for internal use you need more business analysis skills. And product managers will benefit from knowing about business analysts and business analysts will benefit from knowing about product management.

You may also need specialist domain knowledge – you might need to be a subject matter expert in your own right, or you might become an SME in given time.

Some understanding of business strategy, finance, marketing, process analysis and design, user experience design and more.

Don’t underestimate the skills and experience you need to be an effective Product Owner.

  1. 2. Authority

At the very least a Product Owner needs the authority to nominate the work the team are going to do for the next two weeks. They need the authority to choose items form a backlog and ask the team to do them. They need the authority not to have their decisions overridden on a regular basis. (OK, it happens occasionally.)

As a general rule the more authority the Product Owner has the more effective they are going to be in their role.

The organization may confer that authority but the team need to recognise and accept it too.

I’ve seen many Product Owners who while they have the authority to nominate work for a team don’t have the authority to throw things out of the backlog. When the only way for a story to leave the backlog is for it to be developed it is very expensive. This leads to constipated backlogs that are stuffed full of worthless rubbish and where one can’t see the wood for the trees.

If the Product Owner doesn’t have sufficient authority then either they need to borrow some or there is going to be trouble.

  1. 3. Legitimacy

Legitimacy is different from authority. Legitimacy is about being seen as the right person, the bonafide person to exercise authority and do the background work to find out what they need to find out in order to make those decisions.

Legitimacy means the Product Owner can go and meet customers if they want. And it means that they will get their expenses paid.

Legitimacy means that nobody else is trying to fill the Product Owner role or undermine them. In particular it means the team respect the Product Owner and trust them to make the right calls. Most of all they accept that once in a while – hopefully not too often – the Product Owner will have to say “I accept technologically X is the right thing but commercially it must be Y; full ahead and damn the torpedoes.”

It can be hard for a Product Owner to fill their role if the team believe a senior developer – or anyone else – should be managing the backlog and prioritising work to do.

  1. 4. Time

Finally, and probably the most difficult… Product Owners need time to do their work.

They need time to meet customers and reflect on those encounters.

They need time to work-the-backlog, value stories, weed out expired or valueless stories, think about the product vision, talk to stakeholders and more senior people, and then ponder what happens next.

Time to evaluate what has been delivered and see if it is delivering the expected value. Time to understand whether that which has been delivered is generating more or less value than expected. Time to feedback those findings into future work: to recalibrate expected values and priorities, generate more work or invalidate other work.

Product Owners need time to look at competitor products and consider alternatives – if only to steal ideas!

They need time to work with the technical team: have conversations about stories, expand on acceptance criteria, review work in progress, perhaps test completed features and socialise with the team.

They also need time to enhance their own skills and learn more about the domain.

And if they don’t have the time to do this?

Without time they will rush into planning meetings and say “I’ve been so busy, I haven’t looked at the backlog this week, just bear with me while I choose some stories…”

More often than not they will wing-it, they substitute opinion and guesswork instead of solid analysis, facts and data. They overlook competition and fail to listen to the team and other managers.

And O yes, they need time for their own lives and family.

I sometimes think that only Super Human’s need apply for a Product Owner role, or perhaps many Product Owners are set up to fail from day-1. Yet the role is so important.

I plan to explore this topic some more in the next few posts.

Sprint Zero


Starting a new project? A new product? Have a team transitioning to “Agile” (aka Scrum) and wanting to get ready?

Why not try a Sprint Zero? – Why not indeed.

Sprint Zero is a way to get everything ready to start work, to buy the tools you need, set up your databases, perhaps do some architecture review, tell the rest of the organization about what you are doing, maybe write some requirements, get some additional training, etc. etc.

Sprint Zero doesn’t need to happen in a time box, you can take all the time you need, after all, you haven’t become “agile” yet, Sprint Zero will make you Agile.

Good idea?

Please don’t.

Sprint Zero is a get-out-of-jail-free card that pushes really change down the road a little bit further.

The truth is: there is work to do.

Just start “sprinting”. Schedule an iteration to start tomorrow and get going. Do what you plan for the next week, then review and plan again.

Doing anything else delays the change and potentially reduces motivation.

If you want to call your first iteration Zero then fine; you can call it Zero, or One, or Two, or 57 or even -1 if you want, I don’t care. Give it a number and make sure the next iteration adds one to that number.

All those things I just listed – and more – that you need to do to “get started” are themselves work that needs doing. Why not write them on a card and schedule them like you would any other piece of work in an iteration?

If you haven’t worked out what those things are then No problem, try and schedule some real work. When you find you don’t have something (a database instance for example) simply write out a card, schedule it, do it. The things you need to do to “get ready” are actually things you need to do to build something you do want.

My big worry with Sprint Zero (apart from it injecting unnecessary delay and loss of change momentum) is that it makes work. You may think you need to do something in advance and you might be right. But, then again, you might be wrong. You might do something – like install a database – when you could quite easily postpone that work for a few weeks or even skip it entirely.

Similarly some say “How can we start work until we know what we need to do? – we need to gather requirements, write some stories….”

But again: determining what needs doing, requirements gathering, is itself work to be done.

If it is not clear what needs to be done on day-1 then the first thing to do is to do some work to find out what needs to be done. The initial work may well be requirements discover and story writing with only a little bit of product building.

Speculative, early, requirements gathering, may actually increase the amount of work to do because you capture all the things people think you need to do. Once you start to put the product together you may not need those things but once they are listed as part of the work to do – and especially if they have been promised to someone – then not doing them can be problematic.

Anyway, having written this blog I now wonder if I should have bothered. Now I look about a lot of other people have made similar points already. As I have been writing this post I thought I should find a definition of Sprint Zero, so I did a search. Interestingly, most of the articles that came up on page 1 of Google are also critical of Sprint Zero for similar reasons to myself. So maybe Sprint Zero is an idea whose time has already come and gone.

Down with management!


Si: Welcome Peter, thank you for taking the time to attend this annual performance review

Peter: As long as you know I don’t want to be here, performance reviews are pointless, I told that to my manager Roger. I should be at my desk coding.

Si: Yes, I understand your position, in fact many of us sympathise with it, including Roger and myself

Peter: Well, good, does he also agree that management is a waste of time? If we were really following Agile the way we should then Roger wouldn’t be getting in the way all the time and I could write more code.

Si: Yes, well, we’ve had that feedback from other people and the company is committed to change. In fact, one of the things I wanted to speak to you about today was to tell you Roger is stepping aside.

Peter: What?

Si: Yes, Roger is undertaking retraining and will become a Product Specialist, he himself rejected the title Product Manager because he doesn’t see the need for Managers and even refused the title Product Owner because he doesn’t feel he can “own” something that is a community effort and is ultimately owned by the company shareholders.

Peter: Wow… he has been listening to me?

Si: Yes, in my long conversations with him he has spoken many times of the points you have been making. Of course as part of the management team it wouldn’t have been right for him to speak about his changing views too publicly lest people wonder what was happening.

One of the things I wanted to discuss with you today was how we go about distributing Roger’s other responsibilities. As one the longest serving employees Roger suggested you would have the best insights.

Peter: Right, erh, so what responsibilities are these? – apart from asking me “how long will it take?”

Si: Well some relate to the product. As Product Specialist he will continue to support sales staff making customer visits, he will continue administer the backlog, prioritise work in the planning meeting and draw-up the product roadmap – although he wants to involve more people and change the concept of a roadmap. He plans to increase the number of customer visits, competitor analysis and market scanning activities he has been doing already.

Peter: Good, we need a proper product owner, he was doing the role anyway but didn’t have enough time.

Si: Well yes, in fact he has also requested that you and the other engineers accompany him on customer visits in future.

Peter: What? – but our customers are everywhere but here!!! Saudi Arabia, Finland, Argentina, USA… it takes days to get to those places, I’m needed here, I need to be coding!

Si: Well I’ll let the two of you come to a decision on that. Right now we need to redistribute Roger’s work.

There are the administration matters, signing off holiday, arranging sabbaticals, maternity/paternity leave; the new monthly check-points replacing annual performance reviews need to be staffed. There is a lot of work around contractors, time-sheets, extensions and terminations, to be honest there are constant political battles over the number of contractors and whether we should be using them at all. Personally, I believe these changes will abolish office politics but some people think I’m being naive.

Anyway, a lot of this work could be for someone like myself in HR…

Peter: But that will slow everything down! HR is never responsive and most of them – unlike yourself – don’t know one end of a code stack from another

Si: Yes, quite right, HR is atrociously understaffed so we take forever to do anything, and many of my colleagues just aren’t close enough to the teams – frankly they don’t understand technology.

For those reasons the company is also closing the HR department. The work will be distributed to the teams. So my colleagues and I are all being made redundant. Some larger teams, such as yours, will be allowed to hire “Talent specialists” to help with recruitment matters and I hope to secure a such a job myself. But that also means members of your team will need to interview and select their own Talent Specialist. I expect you’ll be on the interview panel, I hope you will see the depth of my experience.

Peter: O, but we are all so busy coding, that will slow us down! And we don’t know anything about recruiting HR people. Can’t Roger arrange that before he leaves?

Si: Roger’s re-assignment is immediate, he insisted on it. I suggest you and your colleagues get started on selecting your Talent Specialist as quickly as possible.

Still, many of the things I just listed will not fall under the Talent Specialists remit. A self-managing team really should manage a lot of those things itself. The Talent Specialist will be there for staffing issues, CV filtering, negotiating with recruitment agents, diversity monitoring, visa applications, university recruitment fairs, exit interviews, legal issues, and so on.

Peter: Right, well I’m sure out team can dispense with a lot of the administration, we can trust one another to take holiday responsibly.

Si: Quite so, I’m sure you can organise away a lot of the admin work but there will be a rump.

Peter: Could we bring back the Scrum Masters?

Si: Well the Scrum Masters we had were always at pains to point out they were not managers or administrators – I know they are in some other places. Plus, they said they “did themselves out of a job” in the end and recommended their own removal.

Peter: Arh… I remember, I guess we will share the administration around the team. We can work out a process for that – the same as we do in planning meetings.

Si: Remember as you do that some of your team may need management training

Peter: What?

Si: Management, even administration, has its own on standards, rules, jargon, if members of your team are going to do administration and management work, let alone talk to others doing it then they need to know their CapEx from their OpEx, their 4 Ps from their 5 Forces, front-of-house from back-of-house, what you can and can’t say in an exit interview, a SWOT analysis from a …

Peter: Yes, yes I get it, a lot of mumbo jumbo if you ask me…

Si: All the same, is it fair to throw people in at the deep-end? Shouldn’t we help them learn if they want to?

Moving on next we come to Roger’s role in the hierarchy. Sharing company information down to the team and team information up the chain. Plus annual strategy meetings, quarterly reviews, product portfolio boards, key customer engagement.

Peter: We have Slack and Wiki’s so I’m sure we can manage the information alright.

Si: Arh… well, when we rolled this program out in Norway some of the teams tried just that. In a couple of months most of the team members were complaining about information overload. I believe the quote was “How can you expect me to do any coding when Slack keeps interrupting me?”

Peter: Arhhh.

Si: And the Norwegian executives felt they never got a consistent message from the teams because different people would turn up to quarterly reviews each time while strategy sessions started to resemble a Stackover flow argument. Some customers complained that they no longer had a point of contact or got conflicting messages. And lets not even talk about the auditors.

Peter: Well the thing is, none of us have been trained in Strategy or, what did you call it? Portfolio review?

Si: Good thinking, I guess one of Roger’s strengths was the time he spent studying that on his American MBA. Do you want to arrange a coupe of days training in strategy and portfolio review for your team?

Peter: Erh, I’m not sure I know enough to book a training course, and isn’t that going to take people away from coding?

I mean the whole team, all 10 of us, out for 2 days training? And the people going to attend meetings with other departments and teams?

Si: Yes of course it is, but you will be self-managing. The team won’t have anyone else.

And mentioning the team, there is also another issue you’ve raised yourself in the past. Here we are, two very similar males discussing this. In fact your whole team is quite like the two of us. The team are going to have to challenge themselves to improve their diversity.

Peter: Right…

Look, we’ve been here half-an-hour and haven’t started even reviewing my performance, will this take much longer? I’ve got code to write.

Si: Well, there is quite a bit more of Roger’s work to discuss, plus his boss, Jane, is also stepping aside so we need to reconvene with people from the other teams to allocate her remaining work.

Peter: If we are self-managing, could we decide to employ a specialist to pick up a lot of this work?

Si: I expect you could, you’d need to put some costings together.

Peter: Right, I’ll talk it through with the team and see if we can hire a secretary.

Si: Good idea, although the secretary is going to need more than just typing skills.

Peter: Sure, secretary might be the wrong job title. We need someone with skills who we can trust to make these decisions, someone with experience.

Si: Although, you know, such a person may become a bit of a gate-keeper to the team

Peter: Certainly, thats what we need, we don’t want interruptions!

Si: Well, the problem with a gate keeper is that they decide what goes through the gate and what does not. That gives them a degree of power.

And if someone is picking up all the left over work, making decisions for the team, and controlling access to the team, and information flows into and out of the team…

Peter: Yes, is there a problem? That all sounds good to me, this gate keeper will have the power to defend the team

Si: Well… what I was about to say is such a person might be seen by some as having authority… and how will the team members feel about someone else deciding what is told to others?

Peter: Well, we could review their decisions and tell them what to do

Si: Yes… well, that might be seen as a lacking trust or even micro-manegement…

Xanpan – free eBook now!

Hypothesis: If I offer people a free copy of my Xanpan eBook (suggested price $17.50) then people will sign up to my newsletter list – which carries my blog posts, event lists and other news.

Test: Blog it, Tweet it, etc., wait a month and count the subscribers.

Use the form below and check your mailbox for confirmation and links.

Subscribe to our mailing list

* indicates required

Email Format

Outsourcing banana skins: Warning signs that your supplier isn’t as good as they claim


So you think you want to contract out some development work? – yes, you know this area is full of banana skins to slip on, and you know others have problems but you still want to do it.

And you want to contract it out to an “Agile” development shop?

There are no laws against opening a development shop – a digital agency, an outsourcer, a consultancy, call it what you like. That is the beauty of capitalism, it allows pluralism. The hard part is choosing one that is competent, some outsourcers are pretty awful.

Everyone who can spell “Agile” can claim to be Agile, and most hire a copywriter to give them an online spin so they all end up making the same claims.

Those who are half good can coach their staff in what to say. And in truth, most of them genuinely want to do the best possible job for you – and be agile too. But how can you really tell?

Well… when I’m being cynical I think you can’t. The only way to really tell is to give them some work and see what happens. Of course once you’ve engaged someone you need to be both legally and mentally prepared to walk away.

So to help you here are some warning signs that you have stepped on a banana skin and your supplier isn’t as good as you – and they – want to be. You might even say these are warning signs that they aren’t really Agile.

1) Customer involvement

They don’t want customer involvement. They don’t want your people on site. They claim that your people get in the way. They want to be left alone to do things.

Obviously I’m thinking primarily of actual Customers, Users, Product Owners, Business Analysts and so on. These people should be working closely with the suppliers people. They should have direct contact, they should be discussing stories.

If your supplier isn’t embracing your people and treating them as their own team members something is probably wrong.

The supplier should be challenging your people – after all the suppliers are the experts, if they are simply accepting everything you ask for then something is wrong.

The same is true of other people you might want involved: a consultant or Agile coach should be welcome too. And if you decide to ask a third party to inspect their development then they should be open to this too.

Naturally they should also be open about the code too. After all the code will be yours one day.

2) Regular demonstrations

The supplier should be providing regular demonstrations – “show and tell” – as a very minimum. Every couple of weeks I expect the supplier to show the latest work. You – and your people – should be able to see working software offering more than the previous demo.

If the supplier is NOT providing regular demonstrations then you should be worried. Likewise, if the demonstrations don’t show progress get worried. Most of all, if the supplier doesn’t want to talk about why demonstrations aren’t happening, or how they can show progress then something is wrong.

3) Release, release, release

Show and tell demonstrations are good but the real test is to release to live. Releasing all the way to your live system. You might hide it on an obscure URL that nobody knows, or call it a beta release or something, I don’t care, the closer they can get to real live the better.

You supplier should be releasing to a live environment – or an exact copy – very very regularly.

The longer the supplier goes without an actual release the more nervous you should be. Sure, once in a while things go wrong and nobody is perfect. They may go two weeks with nothing to show for it. I don’t like it – and neither should you – but an occasional gap is OK.

Going four weeks without a release… I suppose it might happen in the early stages of the work. But it is in the early stages that you most need reassurance that they can deliver something – anything!

Six weeks with no release… well here we are into “3 Strikes and you are out.” Sure they will be able to give technical reasons why things messed up three times in a row. But take it from me, something is wrong.

The longer they go without an actual release to more concerned you should be and the more you should offer to work with them to address the issue.

Eight weeks? – eject, eject, eject.

4) Show me the tests

Maybe this should be warning sign #1 but for this you need someone technical, someone who knows what a test looks like. In the show-and-tells your supplier should be able to show you automated tests executing. Not very exciting perhaps and certainly not meaningful to the business but if they can’t show you then how do you know they even exist?

And if your supplier doesn’t have an automated test suite then it is certainly time to get out.

Ultimately the system they are building is yours. Your people will need to maintain it, or you will need to pay someone else to maintain it. Without automated tests that is going to be hard. Skipping tests now might make it look like you are saving money but you are not, even in the short term the lack of tests will bite you hard, it will push up costs and destroy schedules.

5) “Feature complete”

The phrase alone should be a warning sign. Equally the words “75% feature complete” (or any other percentage) is a big red flag.

If the supplier doesn’t have a test suite, if they can’t show working (preferably releasable) code then its probable they are feature stuffing. They will say they are making progress because “60% of features are done”. They may even start to claim they are feature complete but remember: a feature without a test isn’t done.

A feature without a test is pure risk. At any time a defect can put a hand up and say “Fix me!”

An automated test isn’t a guarantee of bug free code but without automated tests then I guarantee you have defects waiting to appear.

If you are in a relationship exhibiting any of these five sign then it is certainly time to talk. It may be time to end. But how do you avoid getting into that position?

Let me be as clear as I can: both you and your supplier should prioritise working, usable, functionality over more functionality. As the old saying goes “A bird in the hand is worth two in the bush”, working, deliverable (even better released) features are the priority, there should be less work in progress, fewer incomplete features, fewer “almost done” and as few as possible defects.

While cynical me thinks you might not be able to avoid it that doesn’t mean you shouldn’t try, so here are four warning signs that you are about to step on a banana skin:

1) “Agile is not that different”

Don’t let them tell you Agile isn’t different. In many ways it isn’t but if a supplier is trying to persuade you that you don’t need to change the way you work then it is a sign that either a) they don’t appreciate the magnitude of the change or b) they will tell you anything to get the contract signed.

Since you want a supplier who will challenge you it might not be a good idea to hire one who doesn’t like challenging you or doesn’t prepare you for difficulties.

2) “We are certified”

An extension of warning sign #1 is that the supplier is proud of how certified they are: ISO-9001, PMI, PRINCE-2, CMMI – some in the Agile world would regard these certifications as warning signs in their own right.

Scrum Master Certified, Agile Project Manager Certified, Scrum Product Owner Certified: these are slightly better but anyone who can’t tell you the flaws in all these certifications has myopia.

Question any organization that offers up badges rather than working products.

(Disclosure: for better or worse I hold a couple of Kanban certifications, while I think Lean Kanban University have done a better job than many in making their certifications meaningful I don’t think they are a panacea either. Anyway, Kanban certifications aren’t as recognised as those I just mentioned.)

3) Fixed or long term contract

IT suppliers have a long history of locking clients into “fixed contracts” – fixed scope, cost and time. These contracts are utterly flawed. Anyone claiming they can fix everything is a charlatan. Give them a copy of “Dear Customer: The Truth about IT” (the Xanpan prologue) and say goodbye.

Similarly locking yourself into a long term contract with a supplier before you have done some work with them is a bad sign. Do a small piece of work, for a small fee, with your potential supplier and see if they are as good as they say.

In my experience the best – most “agile” – digital suppliers can pick and choose their customers. If your supplier needs you more than you need them then it is a bad sign.

Read more? Subscribe to my newsletter – free updates on blog post, insights, events and offers.

Minimally Viable Team in a nutshell


Last week I was in Holland helping a client with their agile adoption and digital transformation. When the subject of teams came up I started talking about Minimally Viable Teams. Yesterday I found myself writing an e-mail to the client expanding on the idea. And it seemed to me that the e-mail – or an edited version – was worth sharing here…

The idea of an Minimally Viable Team (MVT) is based on the observation that if a team is overstaffed then team members will find work for themselves – Parkinson’s Law.

Mix in Conway’s Law: the recognised phenomenon where team copy the organization structure they are in. So for example, if you have a database expert on the team the final design will use a database whether one is needed or not.

If one is aiming for a self-organizing, goal-directed, value-seeking team then making any decisions about the team, the software design, or even the problem before work begins is questionable. The more decisions that are made the more the team is constrained, the more the team is constrained the less it is master of its own destiny.

Further, those decisions made before work begins: one expects them to be rational, which means some pre-work is needed to understand what decisions are needed and make the decisions. That pre-work costs time and money. The more money that is committed then the more difficult (i.e. more career threatening) it becomes to reverse those decisions if things go wrong.

Some companies spend an awfully long time thinking and planning to do something: longer than it takes to actually do the thing. I once visited a company which had spent five years planning a particular project and not building anything.

Add two more things to this.

We know from experience that planning has rapidly diminishing returns. A little bit of planning creates great learning, but after a little while the rate of learning drops off. Very soon learning by doing becomes more effective, i.e. switch from thinking about what might be done to trying to do it.

This has never been truer than today – 2018: with the computing power and tools it is faster and cheaper than ever to build a prototype, a concept, an MVP, version 1, alpha version or whatever else you want to call it.

However, going to the other extreme and doing no pre-work doesn’t make a lot of sense either.

Enter the Minimally Viable Team: the team jumps to doing, all that pre-work is given to the team. They get to decide what is needed.

To traditionalists the team/project/product is launched prematurely but actually all we are doing is extending the start date backwards so that the pre-work is now part of the thing. By tasking the initial team with all the traditional pre-work the team becomes master of their own destiny again. And they can choose to approach the mission with a traditional approach (market research, architecture design, resource planning) or in an agile/digital fashion (build a small product and test it) – that is their choice.

The MVT idea is to “starve” the team and make them pull only the necessary resources as and when they need them. When organizations decide who (which roles) will be on the team in advance they are in effect designing the software.

And since agile approaches and modern tools allow us to make progress that much faster why not move more quickly to a working product? Minimise the design, postpone the architecture.

This approach also means the initial team can be kept small which means they are cheap. So if they conclude the project shouldn’t be done the organizational inertial is less and the project can be cancelled. Which hopefully means the organization will take more chances on more ideas.

Try this thought experiment.

On Day-Zero there is nothing.

Someone decided there should be Product X. How did this happen? They may have had the idea days ago and have spent the intervening time researching the market, the competition, the problem and so on. (During which time their normal job has been disrupted, the sooner they can dedicate themselves to the new idea the faster things will happen.)

On Day-Zero they talk to an architect who considers a design.

This takes a few days at the end of which there is an outline design and the architect suggests the team needs four programmers, two testers, a UXD and a DBA. Plus a project manager and product owner. 10 in all.

It now takes time to make the business case and gather those resources.

At that point work can officially begin, call that D-Day.

Then the team need to learn to work together, build something and launch it into a market.
They also need to understand what the architect had in mind.

Officially the project began on D-Day, or perhaps the day the business case was approved.

How much time has been spent so far? Without testing the market? Allowing competitors to do something? – all this time cost of delay has been at work changing the business case.

How has all that “getting ready” time been accounted for? How has this work been managed?

The MVT approach says: Time is of the essence the team should decide all these things.

So, as quickly as you can, spend a little venture money on an MVT.

That team has to investigate the market, competition, problem, etc. The team can think about architecture but their primary aim is to build something, and MVP, a prototype, a proof of concept, whatever – build it, show it to customers, launch it, put it in the market.

By keeping it small the team can quickly invalidate ideas which don’t work. Ideas that do work can be built on. They are free to learn.

The initial MVT will do all the same things that a “pre project” phase would do but in a much more agile/digital way. The MVT also allows for continuity, when the team find success the team that can be expanded and grown – applying Gall’s Law.

This also looks a lot more like a start-up than a normal corporate project.

If the idea of a Minimally Viable Team is new to you then check out the discussion in Continuous Digital or some of my earlier blog posts on MVTs.

Read more? Subscribe to my newsletter – free updates on blog post, insights, events and offers.

Nature abhors an information void


No. 6: What do you want?
Voice: Information
No. 6:You won’t get it
Voice: By hook or by crook we will

Information… we all want information… Facebook updates, Tweets, 24-hour rolling news, the Donald Trump Big Brother House… the opening scenes and words of The Prisoner continue to echo, Patrick McGoohan and the other writers got it right, they were just 50 years early.

Human beings have insatiable thirst for information – even when we know rationally that information is useless is pointless we still want it. We persuade ourselves that something might be happening that we need to know about.

Just today I was driving when my mobile phone started to ring. It was highly unlikely to be anything but still my mind started to think of important things it could be. I had to stop the car and try to answer it. Of course, it was spam, a junk call, caller-ID told me that so I didn’t answer.

Every one of us has information weaknesses. In part it is dopamine addiction. We may look down on those who watch “vanity metrics” but we all information fetishes whether they be, metrics, scores, “facts” or celebrity gossip.

Whether e-mail, Twitter, Facebook, WhatsApp, SMS, Slack, some other medium or social media we all need information and a dopamine fi. Has only replied to my tweet? Has anyone retweeted my last tweet? Has anyone followed me today?

Sometimes it is impossible to believe that nobody has retweeted my fantastic tweet, or that a potential client hasn’t immediately replied to an e-mail, or that… I’ve even on occasions found myself picking up my phone and going to the mail app when I’ve only just walked away from answering e-mail on my PC – as if the e-mail on my phone is better than the e-mail on my PC!

The only thing worse than having a mailbox full of unanswered e-mails is an empty mailbox – mailbox zero – which stays empty.

Sometimes one demands information when there just isn’t any. I think that is what number 6 really meant when number 2 repeatedly asked him for information: there wasn’t anything more than he had said. He had given his information, if others demanded more then it was simply because they couldn’t accept what they had been told.

I’m sure all parents have experienced children in the back of the car who ask: “Are we there yet?”. To which you reply “No – it will be at least an hour”. And then, five minutes later you hear “Are we there yet?”

And who hasn’t felt the same way about project managers? Or technical leads? Or product managers? product owners? business analysts?

Children don’t stop asking because… well, maybe because they don’t understand the answer, they have a poor concept of time. Or maybe because they really want the answer to be “Yes we are there.” As small people children also want information.

Isn’t that the same when other people ask you the “Have you finished foo yet?” and even “When will it be ready?” While one hopes they have a better concept of time they don’t necessarily take in the answer, and they hope and hope and hope that the answer will soon be the answer they want it to be. People are very bad at handling information voids.

Manager types might dress the question up in terms of “The business needs to know” how often does that disguises the real truth: somebody didn’t like the last answer and is hoping that if the question is posed again the answer might be the one they want.

The project manager who checks in every few hours is no different than the developer who leaves their e-mail open on a second screen, or the tester keeps Twitter in the background. Each of them wants to know information!

Our difficult in dealing with information voids means we constantly search for information. And if we can’t find it we create pseudo-information: time based project plans which purport to show when something will be done or system architecture documents which claim to show how everything will work. Are the project managers and architects who create these documents are just seeking information? Dopamine?

Long time readers may remember my review of time-estimation research. Some of the research I read showed that people in positions of authority, or who claimed expert knowledge, underestimates how long work will take more than the people who do the work. Researchers were not clear as to whether this effect was because those in authority and experts let their desire for the end state influence their time estimation or whether it was because these people lacked an understanding of the work in detail and so ignored complications.

And it is not just time based information. Requirements documents are often an attempt to discern how a system may be used in future. System architecture designs are an attempt to second guess how the future may unfold. Unfortunately, as Peter Drucker said “We have no facts about the future”.

Faced with an information void we fill it with conjecture.

Sadly I also see occasions where the search for answers disables people. Sometimes people search for information and answers which are within their own power to give. Consider the product owner inundated with work requests for their product. They search for someone to tell them what they should do and what they should not do. Faced with an information void they look for answer from others. But sometimes – often? always? – the answer is within: as product owner they have the authority to decide what comes first and what is left undone.

I have become convinced over the years that often people ask for information that simply doesn’t exist. When the information isn’t presented they fill in the blanks themselves, they assume the information does exist and isn’t being shared. In some cases they create conspiracy theories or they accuse others of being secretive. But because of doubt they they don’t act on the information.

It is easy to think of examples in the public eye but I think it also happens inside organizations. Often times managers really don’t know what the future will hold but if they don’t tell people then they are seen as hiding something. If they deny information exists they may be seen as stupid or misleading.

The same happens the other way around, the self same managers – who really don’t know as much as people think they do – ask programmers, testers, analysts, etc. for information which doesn’t exist and which maybe unknowable. Telling your manager “you don’t know” might not be something you feel safe doing, and if you do then they may go and ask someone else.

In almost every organization I visit people tell me “We are not very good at communicating around here.” Again and again people tell me they are not told information they “should” be told. I’ve never visited an organization where people tell me “Communication is great around here” and while I’ve visited places where people say “We have lots of pointless meetings” nobody tells me “We are told too much.”

My working assumption in these cases is simply: The information doesn’t exist.

This is Occam’s razor logic, it is conspiracy free, it doesn’t assume the worst of people. I don’t assume people are keeping information secret – either deliberately or through naive understandings of what other people want.

So, the real answer for No. 6 should be “I’ve told you the truth, maybe you can’t accept it.”

Read more? Subscribe to my newsletter – free updates on blog post, insights, events and offers.

I am guilty of Agile training


Over Christmas I was thinking, reflecting, drinking…

Once upon a time I was asked by a manager to teach his team Agile so the team could become Agile. It went downhill from there…

I turned up at the clients offices to find a room of about 10 people. The manager wasn’t there – shame, he should be in the room to have the conversations with the team. In fact half the developers were missing. This company didn’t allow contractors to attend training sessions.

For agile introduction courses I always try and have a whole team, complete with decision makers, in the room. If you are addressing a specialist topic (say user stories or Cucumber) then its OK to have only the people the topic effects in the room. But I am talking about teams and processes, well I want everyone there!

We did a round of introductions and I learned that the manager, and other managers from the company, had been on a Scrum Master course and instructed the team to be Agile. Actually, the company had decided to be Agile and sent all the managers on Scrum Master courses.

So the omens were bad and then one of the developers said something to the effect:

“I don’t think Agile can help us. We have lots of work to do, we don’t have enough time, we are already struggling, there is masses of technical debt and we can’t cut quality any further. We need more time to do our work not less.”

What scum am I? – I pretend to be all nice but underneath I allow myself to be used as a tool to inflict agile pain on others. No wonder devs hate Agile.

My name is Allan and I provide Agile training and consulting services.
I am guilty of training teams in how to do Agile software development.
I am guilty of offering advice to individuals and teams in a directive format.
I have been employed by managers who want to make their teams agile against the will of the team members.
I have absented myself from teams for weeks, even months and failed to provide deep day-in-day-out coaching.

In my defence I plead mitigating circumstances.

One size does not fit all. The Agile Industrial Complex* has come up with one approach (training, certification and enforcement) and the Agile Hippies another (no-pressure, non-directive, content free, coaching).

I don’t fit into either group. Doing things differently can be lonely … still, I’ve had my successes.

I happen to believe that training team members in “Agile” can be effective. I believe training can help by:

  • Providing time for individuals to learn
  • Sharing the wisdom of one with others
  • Providing the opportunity for teams to learn together and create a shared understanding
  • Providing rehearsal space for teams to practice what the are doing, or hope to do
  • Providing a starting-point – a kick-off or a Kaikaku event – for a reset or change
  • and some other reasons which probably don’t come to mind right now

Yes, when I deliver training I’m teaching people to do something, but that is the least important thing. When I stand up at the start of a training session I image myself as a market stall holder. On my market stall are a set of tools and techniques which those in the room might like to buy: stand-up meetings, planning meeting, stories, velocity, and so on. My job is to both explain these tools and inspire my audience to try. I have a few hours to do that.

As much as I hate to say it, part of my job at this point is Sales. I have to sell Agile. In part I do that by painting a picture of how great the world might be with Agile. I like to think I also give the audience some tools for moving towards that world.

At the end of the time individuals get to decide which, if any, of the tools I’ve set out they want to use. Sometimes these are individual decisions, and sometimes individuals may not pick up any tools for months or years.

On other occasions – when I have time – I let the audience decide what they want to do. Mentally I see myself handing the floor over to the audience to decide what they want to do. In reality this is a team based exercise where the teams decide which tools they want to adopt.

If a team wants to say “No thank you” then so be it.

In my experience teams adopting Agile benefit greatly from having ongoing advice on how they are working. Managers benefit from understanding the team, understanding how their own role changes, and understanding how the organization needs to change over time.

Plus: you cannot cram everything a team need to know into a few hours training and it would be wrong todo so. You don’t want to overload people at the start. There are many things that are better talked about when people have had some experience.

Actually, I tend to believe that there are some parts of Agile which people can only learn first hand. They are – almost – incomprehensible, or unbelievable until one has experience. That is one of the reasons I think managers have trouble gasping agile in full: they are too far removed from the work to experience it first hand.

You see, I believe everyone engages in their own sense making, everyone learns to make sense and meaning in the world themselves. In so much as I have a named educational style it is constructivist. But my philosophy isn’t completely joined up and has some holes, I’m still learning myself.

When I do training I want to give people experiences help them learn. And that continues into the work place after the training.

So I also offer coaching, consulting, advice, call it what you will.

But I don’t like being with the team too much. I prefer to drop in. I believe that people, teams, need space to create their own understanding. If I was there they wouldn’t get that space, they wouldn’t have those experiences, and possibly they wouldn’t take responsibility for their own changes.

One of my fears about having a “Scrum Master” type figure attached to a team is that that person becomes the embodiment of the change. Do people really take responsibility and ownership if there is someone else there to do it?

I prefer to drop in occasionally. Talk to individuals, teams, talk about how things are going. Talk about their experience. Further their sense making process. Do some additional exercises if it helps. Run a retrospective.

And then I disappear. Leave things with them. Let them own it.

Whether technical skills are concerned – principally TDD – it is a little different. Because that is a skill that needs to be learned by practice. I don’t tend to do that so I usually involve one of my associates and they are sometimes embedded with a team for a longer period.

Similarly, I do sometimes become embedded in an organization. I can be there for several days a week for many weeks on end. That usually occurs when the organization is larger, or when the problems are bigger. Even then I want to leave as much control with the teams as I can.

On the one hand I’m a very bad person: I accept unwilling participants on my training courses and then don’t provide the day-to-day coaching that many advocate.

On the other hand: what I do works, I’ve seen it work. Sometimes one can benefit from being challenged, sometimes one needs to open ones mind to new ideas.

If I’m guilty of anything I’m guilty of having a recipe which works differently.

And that team I spoke of to start with?

One day two some people did not return: that was a win. They had worked out that it was not for them and they had taken control. That to me is a success.

Most people did return and at the end, the one who had told me Agile could do nothing for them saw that Agile offered hope. That hope was principally an approach to quality which was diametrically opposite to what he initially thought it was going to be and was probably, although I can’t be sure, the opposite of what his manager thought Agile meant.

It is entirely possible that had his manager been in the room to hear my quality message I’d have been thrown out there and then. And its just possible I might have given him food for thought.

But I will never know. I never heard from them again. Which was a shame, I’d love to know how the story ended. But that is something else: I don’t want to force anyone to work with me, I don’t lock people in. That causes me commercial headaches and sometimes I see people who stop taking the medicine before they are fully recovered but thats what happens when you allow people to exercise free will.

O, one more thing, ad advert, I’m available for hire, if you like the sound of any of that then check out my Agile Training or just get in touch.

*Tongue in cheek, before you flame me, I’ve exaggerated and pandered to stereotypes to effect and humour.

Read more? Subscribe to my newsletter – free updates on blog post, insights, events and offers.

Conclusion: Who works on what – Comparative advantage part 3 of 3


In my last two posts – Who should work on what? part1 and part 2 – I’ve tried to apply the comparative advantage model from economics to the question of which software developer should work on what. The model has come up with two different answers:

  • If productivity (measured by quantity of features is the goal) then it probably makes sense for everyone to work on the product that they are comparatively most productive on (comparatively being the key word here.)
  • If value produced in the goal then it may well make sense for everyone to work on the most valuable features (or product) regardless of personal strengths.

Along the way I’ve highlighted a number of difficulties in applying this model:

  • If common resources are being used, or if doing one piece of work impacts another, then the model doesn’t work.
  • There is no consideration of time or urgency in the model. When urgency enters the picture then productivity may well suffer.
  • Over time things may change: backlogs will stratify and people will learn.
  • Operating this model in practice requires data which is usually unavailable and so getting the data would itself take time.

At this point it is tempting to throw ones hands up in the air and say: “We’ve learned nothing!”

But I don’t think so. I think there are lessons in here.

Right at the start of this I knew this was a difficult question to answer, trying to answer it has shown just how hard it is to get a definitive answer. There are still more assumptions which could be relaxed in this model and still more variables that could be added.

The model has also shown how important it is to have a sense of value. Not only between products but between features. That in turn demonstrates the importance of both valuing work in the backlog and regularly reviewing those valuations.

However, the first big lesson I think that needs learning here is: you have to know what your intention is.

You need to know what you are trying to optimise.
You need a strategy.

For example:

  • Do you want to maximise the quantity of features delivered?
  • Do you want to maximise the value delivered? (probably measured in money)
  • How much do you want to allow for urgent work? And to what standard are you going to hold those requests?
  • Do you want to promote specific knowledge (so one person can become more productive in one domain) or spread knowledge around (so many people can work on many different things)?

In many this is going to be a self-fulfilling prophecy, the result will be what you put in. That is, if people only work on one product then moving people between products will get harder and less productive. If people follow the value then value delivered will increase as people become more productive in the products with the higher value.

Knowing what your intention is should be the first step to formulating a strategy. And having a strategy is important because answering that question – “who should work on what?” – is hard.

To answer that question rationally one needs to create a model, a model far more complex than my model, then calculate every variable in the model – plus keep the variables up to date as they change. Then to apply that model to every work question which arises.


Alternatively one can formulate a rule of thumb, a heuristic, a rough guideline, a “good enough” decision process. This might sound a bit amateurish but as Gerd Gigerenzer says in Risk Savvy:

“To make good decisions in an uncertain world, one has to ignore part of the information, which is exactly what rules of thumb do. Doing so can save time and effort and lead to better decisions.”

To build up such rules of thumb requires experience and reflection, something which might be described as intuition.

So to answer my original question in terms an economist would recognise: It depends.

Read more? Subscribe to my newsletter – free updates on blog post, insights, events and offers.

Adding value – Who works on what? – part 2 of comparative advantage


In my previous post I tried to use the economic theory of comparative advantage to answer the question:

Who should work on what? or Shouldn’t every developer work on the software where they are most productive?

The economic model gave an answer but more importantly it provided a framework for answering the question. As I examined the assumptions behind the model it became clear there are many other considerations which deserve attention.

Perhaps the most important one is: value.

The basic economic model looks, perhaps naively, at quantity of goods produced. Really, one should consider the value of the goods produced. Not only did the model assume that every feature is the same size but it also assumed that all features have the same value.

Flipping back to the basic model, lets assume that each Bonds feature generates $10,000 in revenue while each Equities feature generates $20,000. Now the options are:

  1. Jenny and Joe both work on Equities, they produce seven features and generate $140,000 in revenue.
  2. Jenny and Joe both work on Bonds, they produce seven features and generate $70,000 in revenue.
  3. Joe works on Equities and Jenny on Bonds, the six features they produce generate $80,000 in revenue.
  4. Joe works on Bonds and Jenny on Equities, the eight features they produce generates $130,000 in revenue.

Clearly option #1 is the one to choose because it generates the greatest revenue even though Joe would be more productive if he were to work on Bonds. Adding value to the basic model changes the answer.

Now, again there is an assumption here: all features produce the same value. That is unlikely to be true.

Indeed, over time if no work is done on Bonds it would be reasonable to assume the value of the features would increase. Not that all features would increase in value but failure to do any would mean some of those in the backlog would become more valuable. In addition new requests might arise which may be more valuable than existing requests.

Further, while the value of Bonds features would be increasing the value of Equities might be falling. This follows another economic theory, the law of diminishing marginal utility. This law states that as one consumes more of a given product the added utility (i.e. value) derived from one more unit will be less and less.

So now we have exposed another assumption in the model: the model is static. The model does not consider the effects over time of how things change – I’ll come back to this in another context later too.

Over time the backlogs for both products will stratify, each will contain some items which are higher in value than average and some which are lower in value.

Lets suppose each product has its own backlog:

  • Equities backlog contains seven features with the values: $60,000, $54,000, $48,000, $42,000, $36,000, $30,000 and $24,000.
  • Bonds backlog contains another seven features with the values: $32,500, $10,000, $7,000, $6,000, $5,000, $4,000 and $,3000.

Now there are (at least) four options open:

  1. Equities: both Jenny and Joe work on the equities product. Together they will deliver seven features and a total of $294,000 of value.
  2. Bonds: both Jenny and Joe work on the bonds product. Together they will deliver seven features and a total of $67,500 of value.
  3. Specialise: Jenny does five equities features ($240,000) and Joe three bonds features ($49,500) delivering a total of eight features and $289,500.
  4. Value seeking: Jenny does her five equities features but Joe delivers one bonds feature, one equities feature and gets to go home early. In total they deliver six features and $302,500.


The highest value option if #4, which delivers $13,000 more than if they specialise. That might seem counter intuitive: the option that delivers the most money delivers the least features. And again it shows deciding work in the absence of value can be misleading.

The second best option is for both to do Equities only, this delivers $8,500 more than specialisation. Adding value to the basic model isn’t a big change but it has changed the answer. When output was measured in features then specialisation looked to be the best option.

Returning to the question of the static model, there is one more assumption to relax: Learning. Economist J.K.Galbraith pointed out that the comparative advantage neglects to factor in learning, and I’ve done the same thing so far.

Assuming Joe specialises in Bonds and spends most of his time working there he will learn and in time he will become more productive. Suppose after a year he can produce 5 bonds features in the time he takes to produce 2 equities features – a 66% improvement.

Now how to the numbers stack up? What is the revenue maximising choice now?

And perhaps more importantly, how long would it take before Joe’s increased output paid for all the time he spent learning?

But, another what-if, what if Joe had specialised in Equities instead? He would now be more productive on a product with higher value features.

Again the question “Who should work on what?” needs to consider intent. Which product do you want Joe to learn? Which product is expected to have the highest value? Are you maximising value or quantity?

As usual, you can argue with my model and question my assumptions but I think that only demonstrates my point: these things need thinking about.

If you want you can continue relaxing the assumptions and do more what-if calculations – for example I’ve assumed Jenny and Joe cost the same. Nor have I factored in risk or cost-of-delay. This model can get a lot more complicated. I’ve also assumed that partially done features have no value at all, each week starts afresh and no work carries over.

Read more? Subscribe to my newsletter – free updates on blog post, insights, events and offers.