Consider this a gift, its also an experiment. Numbers are limited so if you would like to join please e-mail me today – if it goes well I’ll repeat, although I might ask for money next time.
I’m going to tun an online workshop entitled: Stories and Value.
Participation is limited to a 16 and its going to be first come first served – blog/newsletter readers are getting the first chance to sign-up.
This is based on my existing “Requirements, Backlogs and User Stories” workshop which has itself mutated into a discussion of stories and value. The workshop will run as a series of 90 minute sessions, one a week for four weeks online.
I want the workshop sessions to remain interactive, I’m sure I will use some slides at some point but I want to keep it interactive. So I’m going to limit participation to 12.
The draft schedule is:
Workshop 1: How value influences our thinking
Workshop 2: Good and Bad User Stories
Workshop 3: Estimating story value
Workshop 4: Time value profiles and closing discussion
I plan on using exercises in throughout and I think I know how to run them online. And I want discussion! – I may even set a little homework between sessions.
But in all honesty, it’s an experiment. So, I’m not planning on charging for this – it is Free!
If you find it valuable you can make a payment – like those “pay what you like” restaurants. That will itself be feedback.
I’m thinking Wednesday, 3pm UK time so those in mainline Europe could join too (sorry US and Asia, maybe next time); on a Zoom conference. Start next week, April 1 ? – once I know who’s in we might debate this between ourselves.
My thinking is still developing on this so let me know if you have any ideas to contribute. (And if you can’t join but want to let me know, feedback is valuable too! Likewise, if you are tempted but want to see something different please tell me and I’ll see what I can do.)
Regular expressions are widely used, but until recently they were rarely studied empirically (i.e., just theory research).
This week I discovered two groups studying regular expression usage in source code. The VTLeeLab has various papers analysing 500K distinct regular expressions, from programs written in eight languages and StackOverflow; Carl Chapman and Peipei Wang have been looking at testing of regular expressions, and also ran an interesting experiment (I will write about this when I have decoded the data).
Regular expressions are interesting, in that their use is likely to be purely driven by an application requirement; the use of an integer literals may be driven by internal housekeeping requirements. The number of times the same regular expression appears in source code provides an insight (I claim) into the number of times different programs are having to solve the same application problem.
The data made available by the VTLeeLab group provides lots of information about each distinct regular expression, but not a count of occurrences in source. My email request for count data received a reply from James Davis within the hour
The plot below (code+data; crates.io has not been included because the number of regexs extracted is much smaller than the other repos) shows the number of unique patterns (y-axis) against the number of identical occurrences of each unique pattern (x-axis), e.g., far left shows number of distinct patterns that occurred once, then the number of distinct patterns that each occur twice, etc; colors show the repositories (language) from which the source was obtained (to extract the regexs), and lines are fitted regression models of the form: , where: is driven by the total amount of source processed and the frequency of occurrence of regexs in source, and is the rate at which duplicates occur.
So most patterns occur once, and a few patterns occur lots of times (there is a long tail off to the unplotted right).
The following table shows values of for the various repositories (languages):
The lower (i.e., closer to zero) the value of , the more often the same regex will appear.
The values are in the region of -2.5, with two exceptions; why might StackOverflow and npm be different? I can imagine lots of duplicates on StackOverflow, but npm (I’m not really familiar with this package ecosystem).
I am pleased to see such good regression fits, and close power law exponents (I would have been happy with an exponential fit, or any other equation; I am interested in a consistent pattern across languages, not the pattern itself).
Some of the code is likely to be cloned, i.e., cut-and-pasted from a function in another package/program. Copy rates as high as 70% have been found. In this case, I don’t think cloned code matters. If a particular regex is needed, what difference does it make whether the code was cloned or written from scratch?
If the same regex appears in source because of the same application requirement, the number of reuses should be correlated across languages (unless different languages are being used to solve different kinds of problems). The plot below shows the correlation between number of occurrences of distinct regexs, for each pair of languages (or rather repos for particular languages; top left is StackOverflow).
Why is there a mix of strong and weakly correlated pairs? Is it because similar application problems tend to be solved using different languages? Or perhaps there are different habits for cut-and-pasted source for developers using different repositories (which will cause some patterns to occur more often, but not others, and have an impact on correlation but not the regression fit).
Today I’m giving away my crown jewels. Once you have read this post I can’t run my favourite exercise with you. There is a bit of experiential learning I can’t give you. So if you’ve rather have the experience stop reading and go and book yourself on my May workshop.
I’m describing an exercise that models what happens in “the real world(tm).” Plenty of the people who have done the exercise recognise it was a real life problem. The insights are many, and some a little disturbing.
Dozens of teams and the answers are always the same. I live in dread that someone will guess and ruin the exercise but it never happens. Now I’m telling the world that might change.
On screen I put a story something like:
As a widget maker I want an online store to sell my widgets so that I can make money
I separate the room into teams. Each team represents a technology supplier – an agency, an outsourcer, whatever. I want each team to competitively bid on a piece of work. Each team gets to think about the problem and estimate the work. At the end I want each team to be ready to name their price, how long it will take and how many people they need. They may add any more details they like, e.g. staging, design, technology, etc. (and most do).
The teams on the right get a story which says:
As an international widget maker I want to sell direct to customers so that I can cut out distributors. I anticipate $10million turnover within 3 years and have budgeted $1.2m for this project.
15 minutes later the teams on the right read out their bids. They always want a million give or take. They want months, if not years. They want teams of half a dozen or more engineers, testers, UXD, analysts and project managers. They may propose an ongoing maintenance contract too.
Very disconcerting for these teams is that while they are answering and taking questions the other teams, those on the left, often burst out laughing – literally – when they hear these proposals.
What neither side knows is that they have different stories. The teams on the left get a story saying:
As an artisan widget maker I want to sell my widgets to customers so that I can give up my day job. When I make $100,000 a year in sales I can live my dream. My accountant tells me a WordPress website will cost $5,000.
These teams want a week or two, an engineer or two and perhaps $10,000 tops. In some cases they say “We can do it this afternoon, we’ll set up Etsy.” Even if they don’t want to use Esty or eBay they probably propose an OpenSource solution.
So what do you think?
True, it is a semi-artificial set-up but I would argue that these situations happen all the time in “the real world” work environment. However they are usually well disguised and hard to see. Maybe now you will spot them.
That aside there are many potential lessons this exercise illustrates, almost everyone is worth a discussion in its own right. To keep things brief I’ll just highlight some of them:
Anchoring (cognitive bias): individuals are anchored to those numbers, bigger number lead them to frame their answers as bigger numbers.
Assumptions: people jump to assumptions, people automatically fill in the blanks when they lack information and the information they fill in flows from the numbers mentioned. Few questions get asked.
Different solutions: the key lesson for me, confronted with similar problems, people (especially engineers) are capable of formulating very different solutions. Those solutions have different time and cost implications.
Problem bounding: presenting the same problem with different bounding constraints results in massively different solutions.
Effort estimates, and therefore cost estimates, flow from value: whether through anchoring assumptions or solution designs the estimates (time and money) flow from the value available NOT the other way around.
Prior experience often goes out the window. In one run a low-end teams told me: “We did this last week. A digital consultant showed us how to install WordPress and Magento for online retail in the morning and in the afternoon we did it ourselves.” While this team came up with a low cost proposal their colleagues who were given the $1m story forgot everything they learned last week.
People don’t ask questions: I answer questions while teams are creating their answers but people rarely challenge what is asked for. Maybe its because I’m usually in some position of authority as a consultant or workshop trainer and my word should be followed.
Occasionally a team with the million dollar story say “We could do this with eBay/WordPress/Shopify.” I keep a poker face and let them be. Inevitably left alone for long enough they talk themselves into a much more complex and expensive bid.
In fact, the longer I give teams the higher the estimates go. I heard a team in Australia say three times “Those estimates look low, lets double them.” And they did. (Again, planning has diminishing returns.)
So far nobody has offered two solutions: you could offer up a Shopify solution and a custom build solution but nobody has.
While we are going through the exercise the minimal viable product idea often gets mentioned – usually by the teams on the right. So recently I introduced a third story. This read the same as the international widget maker but has an extra paragraph underneath:
MacAllan consulting has advised the company to take an iterative and agile approach to this work using a minimally viable product model.
How do you think teams respond?
Think for a minute… your answer is?
It makes no difference.
Or rather, so far I’ve not had any of the million dollar teams propose anything close to the $5,000 solution. In one case a team with the MVP story actually came in more expensive – and longer – than the million dollar team without the MVP clause.
My learning here: talking MVP makes no difference. If you want an MVP you have to impose constraints. (Hence try an MVT.)
People continue to fill in the blanks after the charade is exposed. I’ve heard software architects argue forcefully they these are different problems because of the money involved and therefore require different architecture. They clearly feel cheated and want to justify the proposal they have made. I suspect they feel I’ve made them look silly and want to undo that impression, I’m sorry if I’ve made anyone feel silly.
I wonder how often that happens in the work place? How many of us would back down in real life? I’d like to think I would but I would probably first try and justify my position.
The architects have a point, in many ways the stories are functionally the same but the differences lie in the non-functional requirements: load, throughput, security and so on. But all of that is inferred by people from the price tag without question.
It makes me said that teams ask so few questions. People easily see themselves as a tailor not as a consultant (my Tailor or Image consultant post.)
Then there are the questions about the bidding process and companies bidding on the work.
Imagine you are the buyer: one supplier bids really low, the others were much higher but inline with your expectations. Would you trust the low bid? Have they blow their credibility?
And as a bidder: if you know the client plans to spend $1,000,000 why bid lower? The engineers cost estimates and designs aren’t relevant. Ideally you bid just below the competition. You are the lowest price with all the credibility and maximum revenue.
For that matter, should you be bidding on this at all?
If you work for a small e-commerce provider in rural Cornwall you may never know about, let alone, bid on a million dollar piece of work from an American multi-national. And if you did, would anyone take you seriously?
Suppose you got your big break: you walk in and offer a quick, low cost solution based on something like Shopify. Would they take you seriously? Would they want to listen?
Do corporations increase their own costs simply by being?
Conversely, if you work for a big consultancy and bid on million dollar deals every week will you be interested in bidding on a $5,000 piece of work? Of course not!
But that also means that if a corporation approaches you for a million dollar online shop, even if you could deliver it in a week’s time running on a third party platform do you have any incentive?
I don’t have answers to these questions. Indeed, there aren’t any right answers. All answers are valid, just some answers are “better” in some places than others.
Ultimately the lesson I take away from this is: we craft solutions within constraints.
More specifically: Engineers engineer within constraints, that is what engineers do.
That reinforces my belief that one needs to really understand benefit (value) and how that changes with time. From there we can work back to a solution.
At the end of January I’m running my 1-day Requirements, User Stories and Backlogs workshop in London with Learning Connexions. I get great feedback from people who attend the course, perhaps because it is mostly exercised based.
If your interested check out the Learning Connexions page – its just one day and won’t break the bank! Hope to see you there.
I feel compelled to write this blog because I keep coming across the wrong type of Product Owner. I feel bad about writing this blog because a) I’ve made these points before in other forums so I’m repeating myself, and b) at the end of the day you, your team, and your organization, is free to define and use any title you like for any role you like, you are free to define any given role as you like.
So let me set out my model of a Product Owner and then at least there is a model to compare any other definition with.
Our old friend the Triangle of Constraints can help here – also know as “The Iron Triangle” and pictured above (I like to call it the FRT triangle). Now notice the version I use is slightly different from the more common model:
Rather than “cost” I label one side of the triangle “People”. I could label it resources but in software development resources are overwhelmingly people and the knowledge they bring. People deserve respect, calling them “resources” makes them sound like paperclips.
For software development costs are function of how many people you have and how long you have them for: costs = people x time. OK, there are some other “resources” to add to costs, e.g. buying laptops, renting time in the cloud, and so on but these are often themselves a function of the number of people you have. Such costs are a small increment on top of the wage bill.
Now the number of people you have is fixed in the short term, or to be more accurate: it is upward fixed. People can get ill or resign at anytime but adding people takes time. So in the short run one can consider that dimension fixed.
Time is also fixed. There is usually a business deadline, or rather a business benefit which is time elastic so you have a date to aim for. And on agile teams there are sprint deadlines (two-weeks, two-weeks, two weeks). So a large part time is fixed.
The final side of the triangle is labelled features or functionality, but might be labelled “requirements”, “the what” or “what are we building” – I like to think of it as the demand side.
With me so far? – so far that should be uncontroversial.
Now the traditional Project Manager role, and to a lesser degree the newer Delivery Manager role, tend to regard the third side – the what side – as fixed. There is a thing to be delivered. It is a known thing. It has been decided on and the manager’s job is to get it delivered.
To this end Project Managers are trained to regard the “thing to be built” as a given, preferably fixed, thing. Their training centres on the other sides: cost and time. They are trained both in rationing these commodities and allocating them in an efficient way. When things go wrong these managers ask for more time (which means more money because the same people need paying) or more people (which both costs more and makes things worse because of Brook’s Law).
So to summarise: traditional Project Managers focus on “when” and the input variables: people/resources and money.
Can you guess what I’m going to say next?
Product Owners – plus Product Managers and Business Analysts – focus on the “what”. What do we need to build next? What has the most benefit? What should we be building for the future?
For Product Owners the time and people are fixed. (This is most obvious in an agile environment but is actually true everywhere sooner or later.)
The thing being built is negotiable, the desired outcome may be achieved by different routes, different technologies and different solutions – the different time and cost will be a consideration but outcome is the primary focus.
In other words: Product Owners are all about the what.
In order to operate in the what-space product owners need authority and legitimacy to flex what they are building. When they don’t have that they are reduced to backlog administrators simply ordering the backlog and feeding it to technical teams. That turns the role into a type of Project or Delivery Manager.
So if you need to tell a real Product Owner from all the other misinterpretations of the role ask:
Does the product owner focus on what?
Can the product owner discuss different solutions and approaches to achieve an outcome?
Is the PO flexible about the backlog? (as opposed to slavishly trying to deliver it all)
Real product owners can answer Yes to all three.
(Notice I’m deliberately being careful in what I say about “Delivery Managers.” This role is still emerging and as such its wrong to generalise about it too much. In so much as a Delivery Manager brings management skills, communication and organization to an effort it can be a positive role. When a Delivery Manager is relabelling of the Project Manager role it can be damaging.)
Now that said, the fact that some organizations choose to define the “Product Owner” role as a role closer to “Project Manager” or “Delivery Manager” rather than a role closer to “Product Manager”, “Business Analyst” or (heaven forbid) business owner causes a lot of confusion.
Perhaps I’m wrong here, perhaps the “Product Owner” is a type of “delivery manager” but I think the majority of writers, thinkers and practitioners agree with me.
Even if you disagree with me I hope we can agree on one thing: because there are different interpretations and implementations of the role there is room for confusion; and that confusion makes it harder to fill the role and harder to be seen as a successful Product Owner.
Like this post? – Like to receive these posts by e-mail?
Is the product owner role impossible to fill well?
Do we set product owners up to fail?
Have you ever worked with a really excellent product owner? Someone you would be eager to work with again?
The lack of really outstanding product owners isn’t the fault of the individuals. I think product owners are asked to do a difficult job and are not supported the way they should be. Worse still, in many organizations the role of product owners is misunderstood, they are seen as a type of delivery manager when in fact they are a type of product owner.
There questions have been on my mind for a while, next month I’m giving a new presentation I’m Oredev in Malmo – and which coincides perfectly with the publication of my new book The Art of Agile Product Ownership (funny that). So by way of preview…
I’ve long argued that product owners need four things in order to do the job well: skills, authority, legitimacy and time. Lets look at each in turn:
1. Skills: the kind of thing a product owner learns on a Certified Scrum Product Owner course are table stakes. Yes POs need to be able to write user stories, split stories, write acceptance criteria, understand agile and scrum, work with teams, plan a little and so on. While necessary such skills are not sufficient.
The bigger question is:
How does a product owner know what they need to know in order to do these things?
How do they know what customers want?
How do they know what will make a difference?
Product owners need more skills. Some POs deliver products which must sell in the market to customers who have a choice. Such POs need to be able to identify customers, segment customers and markets, interview customers, analyse data, understand markets, monitor competitors and much more. In short they need the skills of a product manager.
Other POs work with internal customers who don’t have a choice over what product they use, here the PO needs other skills: stakeholder identification and management, business and process analysis, user observation and interviewing, they need to be aware of company politics and able to manage up. In other words, they need the skills of a business analyst.
And all POs need knowledge of their product domain. Many POs are POs because they are in fact subject matter experts.
That is a lot of skills for any one person. How many product owners have the right skills mix? And if they don’t, how many of them get the training they need?
2. Authority: Product owners need at least the authority to walk in to a planning meeting and state the work they would like done in the next two weeks. They need the authority to set this work without being contradicted by some other person, they need the authority to visit customers and get their expenses paid without having to provide a lengthy explation every time.
3. Legitimacy: Product owners need to be seen as the right person to set the priorities. The right person to visit customers, the right person to agree plans and write roadmaps. They need to be seen as the right person by the organisation, by peers and, most importantly, by the development team.
Authority and legitimacy are closely related but they are not the same thing. While the product owner needs both the lack of either results in the same problem: people don’t take their work seriously and other people try to set the agenda on what to build.
Unfortunately Scrum contains a seldom noticed problem here: product owners are team members, they are peers; the team are self organising and are responsible for delivering the product. (There is an egalitarian ethos even if this is only Implicit.)
But Scrum sets the PO as the one, and only one, who can tell he team what to do.
There is a contradiction.
4. Time: Product owners need time to do their work – which is a lot, just read that skills list and think about what the PO should be doing. And don’t forget the PO is a human being who needs to sleep for seven or eight hours a night, may well have a family and a home to go to.
When does the product owner get to do all of this?
Leave aside the question of where you find such people, or whether our companies pay them enough and ask yourself: do product owners get the support they need from their companies and teams?
So often the PO ends up in conflict with the company about what will be built and when it will be delivered, and they end up in conflict with their team about… well much the same issues every planning meeting.
Think about it: do we ask too much from our product owners?
Do we set up product owners to fail?
I’d love to hear your opinions, comment on this post or drop me a note or leave a comment.
As regular readers might know I’m working on a book called The Art of Product Ownership to be published by Apress later this year. One of the chapters is entitled “Why have a Product Owner” and a few days ago a bunch of ideas crystallised into this…
The aim of the Product Owner is to increase, even maximise, the business value delivered by the team as a whole. The Product Owner does not so much create value themselves as increase the value created by others.
Think of it like this: if the team randomly selected work to do and delivered it to customers then some value would be created. (For the moment I’ll ignore the scenario where that work detracts from the existing value.) The aim of the PO is to ensure the work done creates more value than a simple random selection. The greater the difference, or delta to use a mathematical term, between random selection and an informed selection the better.
The general hypothesis is that intelligent selection of work by a skilled Product Owner will result in both more value being delivered and an increasing delta between intelligent PO selected work and randomly selected work.
This difference the value added by a Product Owner. I like to call this difference the Product Owner Delta.
Now in real life work is seldom randomly so Product Owners are not competing against random selection. In some cases the alternative to a designated Product Owners is someone else: a senior developer, an architect, a manager or someone else. In such cases this person is taking on the Product Owner role. They may not have the title, the aptitude, the skills or official position but when work is selected by one person they are de facto the Product Owner.
In other cases the alternative to the PO might be selection by consensus on the team, or a sub-set of the team. Now it is entirely possible that such a group could outperform a single Product Owner in selecting work – especially is they have market and customer knowledge, some analysis skills, time to do the background research and so on. In some cases this works, for example think of a small start-up staffed by software developers creating software development tools.
However, in some cases selection by committee might be inferior to a random selection. Imagine a team which has never met a customer, argue about what to do, duck key decisions and never say No to any request. Its easy to image a dysfunctional selection committee.
There is more to increasing the Product Owner Delta than simply selecting the highest value items. Timely selection can help too. If decisions are not being made, or committees are spending a long time making decisions then having one person simply make those decisions in an efficient, timely, manner can increase the delta.
Time has another role. Because of cost-of-delay simply selecting the highest value items at any one point in time does not maximise the value delivered. Time Value Profiles (see Little Book of User Stories or my presentations on value “How much? When?”) expose this and need to be another tool in the Product Owners repertoire.
And of course, the Product Owner Delta is not the only reason to have a Product Owner in the team, but it is probably the main reason.
Like this post? – Like to receive these posts by e-mail?
Recently I’ve been looking again at Jobs to be Done and OKRs (Objectives and Key Results). I increasingly see them as story generators and a potential solution to the tyranny of the backlog I described last time.
When I first looked at Jobs to be Done (and OKRs actually) I wondered if they constituted a fourth, top, level on top of Epics, Stories and Tasks. I’ve long argued against having more than three levels of things to do (or requirements as we used to call them.) There are big meaningful things to do (stories), really big things which we don’t as yet understand but look really valuable (epics) and the immediate small things to do right now (tasks).
Actually, I’d rather think most things can be dealt with by two levels and one level is the even better. So adding a fourth “even bigger” thing on top of Epics just felt wrong. Technologists (like myself) have a tendency to map everything into hierarchies; inverted trees with fractal like branches. But not everything is, or should be, a hierarchy, mapping the world into a tree like structure can add complications.
Unlike stories (and epics and tasks) Jobs to be Done don’t really lend themselves to the transactional “Done”. While you could put a Job all the way to Done on your Kanban board and track it from “To do” to “Done” in reality the customer job still exists. Sure you’ve improved it but you can improve it again – another example of Stable Intermediate Forms. This seems to be the great potential of Jobs to be Done, they keep on giving: as much as you improve your product to help with the job you can still improve it some more.
So each time you analyse the Job to be Done you should be able to find more stories to deliver to improve it. Hence the Job to be Done is not a “story” to do, it is a Story Generator. Every time you look at the job to be done you find more stories, every time you examine the result of the latest improvement you find more stories. The job will never be done. Some might see that as a bad thing but that also means the job presents a stable focus for ongoing work.
The same might be true of OKRs but in a slightly different way. Because the objective is reviewed periodically – every quarter or so – it lacks the continuity of Jobs to be Done but perhaps allows the team to switch targets, maybe it is stable enough.
The key results may well be stories in their own right, or they may be things which lead to stories. Either way one can expect some key results to be achieved and marked as done regularly. As they fall they are either replaced by new key results building towards the objective (which themselves lead to stories) or new key results are added for new objectives.
I’m sure there are other story generators out there but the key thing for me is not the mechanism but the existence of the generator. Once you have a story generator you do not need a big backlog of things to do. The generator will replenish the backlog whenever you need more stories – either because you have done them or the value has fallen.
Using a generator removes the need to have a big backlog which removes the tyranny of the backlog. The team are now free(r) to concentrate on delivering value towards their objective.
Finally, I wonder if anyone has used both OKRs and Jobs to be Done together? Right now they feel like alternative generators to me, having both seems like a bit like overkill. Although I accept that maybe OKRs are more corporate and Jobs to be Done are more product focused. Anyone got any experience using them together?
Like this post? – Like to receive these posts by e-mail?
The backlog is a great idea: all the things we think the team will build, or perhaps: things they might build, and it might contain other work, like evaluation or reviews. Yes, the backlog is a great idea, all the stuff the team might do, well perhaps not all, it is seldom complete, after all, as they say “stuff happens”.
The truth is: backlogs have a tendency to grow. All too often I find teams who are struggling under the weight of their backlog. They can’t spare time to do experiments or learn something because there is stuff to do. The backlog becomes a tyrannical ruler and all of it MUST BE DONE.
Look at that hypothetical burn-down chart above. By sprint 15 the team is well on its way to completing all the original work. But the amount of work they need to do is higher than when they started. It is not as if the team have been doing nothing. Look at the next chart, it shows how, most weeks, more work is added than is done.
To my mind finding more work isn’t a problem, indeed teams should be finding more work. Problems stem from the fact that backlogs – and tracking mechanisms like burn-down charts – try to full-fill competing needs.
The backlog is used as a store of ideas for work to do. This makes sense, you can’t do everything today so postpone some to the future. A backlog allows you t move some things from peak time to off-peak, although software development teams rarely seem to see off-peak time.
Plus, having a backlog makes it easier to say:
“Thanks for your suggestion Fred, I’ve added it to the backlog, I’ll let you know when we get to it.”
Rather than:
“Thanks for your e-mail Fred, we deleted it once we stopped laughing.”
It makes sense to give a new idea a quick once-over. But doing a proper analysis is time consuming:Discussing what is being asked for takes time, as does setting acceptance criteria are. And then there is business value to assess and other work priorities to consider. Therefore, put it in the backlog and do all that later (if it ever gets scheduled.)
Without a backlog we would be forced to make a binary decision: do it and do it now or reject it.
In fact the backlog can become a natural filter: as stories age in the backlog some items will jexpire. Unfortunately many Product Owners don’t feel they have the authority to delete old requests so the backlog grows and grows.
I call such a “constipated backlog”: work goes in but very little comes out. When the only way for items to leave the backlog is by doing them the rate of return falls.
The backlog fills another role because so many teams are still expected to meet project success criteria which ask for “everything to be done.” The backlog becomes a tyrant when people believe that one day it will all be done. Worse still, some people plan using this assumption.
People want to know when “it” will be done, how long it will take and how much it will cost. It takes time to answer those questions and if the backlog is growing any answer is going to be wrong.
In fact, it is probably wrong to think everything will ever be done. Unless one freezes the backlog and refuses to add new work then it is likely that low value items will be postponed while new, more valuable items, will take priority.
As an industry we need to drop the idea that a backlog will ever be done: the backlog as repository of ideas is at odds with the backlog as a measure of completeness.
Think about it this way: some of the items in the backlog are very valuable. Some items are worth very little. Some will cost more effort than the benefit they bring. If we do everything then the low value and the high value will all get done. Conversely, if we encourage new ideas and weed-out as many low value items as possible our rate of return will be higher.
But very few teams follow this model. Many more teams are slaves to the backlog, and their quest for an empty backlog is doomed.
A quick update: most of my recent blogs about the product owner role together with some new material, is now available in book form from LeanPub – https://leanpub.com/productownership.
I’m surprised to find I’ve written over 60 pages so far! Still, this is very much a work in progress, there are a few more chapters to add to part 1: The Product Owner role.
But it is part 2 which I’m itching to start writing: the tools of the trade.
For those who don’t know, the beauty of LeanPub is that you can buy my unfinished book now and you will receive updates – to your iPad, Kindle, PC, whatever – as they are produced.
That means three things to me.
Firstly I can receive your feedback – what do you like? What did I get wrong? What else should be in there?
Second, money is feedback, the more of you who buy the book the more motivated I am to write it – I like seeing sales, it tells me people want this book. And if you don’t buy… well maybe I should pivot and abandon it.
Third, it gives me a little beer money.
The bad news is: you also get my dyslexic spelling and grammar.
, where: 
is driven by the total amount of source processed and the frequency of occurrence of regexs in source, and 
is the rate at which duplicates occur.
