Skip to main content

Me and My Bestimates

As Test Manager I fit my team's work into multiple overlapping project schedules which are not under my control. The schedules have, as you might expect, multiple constraints that operate simultaneously, such as:
  • end dates or other milestones
  • dependencies on other parts of the schedule or other schedules
  • level of effort we are able or prepared to commit to different tasks in absolute terms (e.g. contractually) or in relative terms (e.g. based on perceived risk)
  • methodology; different teams in Linguamatics operate different development methodologies, and we work with non-development teams too
  • desired quality level (whatever that means in any given case)
Scheduling in this environment is a challenge even without the wildcard that is the unknowns; the stuff you find out as you go and the contexts that change under your feet.

To do my best to provide the service my (most often internal) clients want - which generally includes some kind of estimate, even if there is little at times to base it on or anchor it to - I think I need a few things, including some model of:
  • the resources that I have or will have available
  • the status of the projects that are ongoing
  • the relative importance of projects to one another
To best provide a framework for my team I think I need a few more things, including:
  • transparency of decision-making
  • clarity of current priorities
  • visibility of actions aimed at improvement
In an attempt to achieve these ends (amongst others) I maintain a spreadsheet that records pieces of work over a certain granularity. The definition of granularity isn't very scientific: if a project feels "worth tracking" then it gets in; if a tester feels they'd like it tracked then it gets in; if it has the feeling of something that might blow up then it gets in. I update and publish this to an internally public location every week, with a short summary of significant changes and various charts showing how we're spending time in various dimensions which I use to analyse how we're working and what changes we might experiment with. (I am a practitioner of and believer in Open Notebook Testing and openness and responsiveness generally.)

For most projects that we decide to track, the tester or testers on a project, and/or me, will pick some initial budget. (There are some exceptions which I'll touch on later.) The budget is based on all sorts of factors, such as those mentioned above and those summarised by George Dinwiddie in Estimating in Comparison to Past Experience  and attempts to take uncertainties into account.

I view the budget as a kind of "top-down" number. It (a) fits into and influences the global scheduling and (b) provides implicit guidance to the testers about the time they have available to test given what we know now. It is part of the project context for them, and should constrain testing in the way that other aspects of that context do.

As work proceeds, testers report the time spent on each project along with their current estimate for the remaining work. I view this as a "bottom-up" number. It (a) invites the tester to think at a level higher than immediate action, to step back, to defocus, to prioritise and (b) provides feedback to me and the bigger scheduling headache.

Some projects are more top-down: for example a piece of work which is set up in a time-box. Some projects are open-ended and bottom-up: for example an investigation into a live support issue. Most projects sit somewhere in between. Projects which are entirely bottom-up will often start with a phase where no explicit budget is set and no estimate is made; instead we'll just record time spent. At some point we'll either reach the end of the investigation, the end of the availability of the tester (in which cases work stops) or some point where we feel we understand enough to set a budget.

Because many of our projects tend to be more schedule-bound, our default position tends to be to assume that project work consumes the budget and then stops and that the tester will reprioritise based on this as necessary. However, frequently something happens on a project that affects the tester's estimate - the feature turns out to be more solid or less complex than we expected; we were able to do more work in unit testing than we anticipated; there's some horrible interaction with another piece of our application that no-one expected. In this case, we will talk about the current budget and the differences between it and the new estimate and what impacts there would be of changing the budget, or not. Some for-instances:
  • the tester discovers that there's some area we didn't think needed testing but actually does and increases their estimate to accommodate it. I might decide, on a schedule basis, that more time is not available, so the budget does not change, and the question becomes one of balancing risk inside the existing budget. Future estimates for this project will need to take this decision into account. 
  • the tester discovers that the implementation is missing some behaviour that the company deems is critical. It will require investigation and retest so they increase their estimate to accommodate that work. I agree that this is important and we increase the budget to provide more time. (And I deal with the global scheduling impact.)
  • I find that there is more time in the schedule, perhaps because something else got cancelled. I tell a tester who previously wanted time for some piece of work that is it now available and increase the remaining budget for that work.
  • I am told that we need to complete work on some project earlier than previously understood and I reduce the budget on it, prompting a discussion about how best to proceed (with the usual iron triangle to constrain us).
What's crucial here is that the discussion can be opened at any point, by either side, should circumstances make that necessary or desirable or even just a sensible precaution. And any negotiation can and does include other stakeholders.

The budget and and estimates mutually constrain one another and contribute uncertainties at their different granularities. They mix science and instinct, planning and gut; known and unknown; they're guesstimates. They go well together, they bounce off each other, they influence one another and reflect one another, they're like best mates; they're bestimates you might say, although I never do.

In some sense I think of them as a simple interface between the complexities at the test level and complexities at the management level. Compressed down into a single number they naturally lose information but, when more information doesn't need to be shared (essentially we are on-track relative to the most recent conversation), that is sufficient and indeed efficient. Having a communication channel open for when more needs to be shared is, as I've just said, crucial. And I'll say it again: available channels of communication are absolutely crucial.


The approach described is a high-level sketch of the process that has evolved over time to try to manage competing constraints and requirements of the kinds that I mentioned at the top. And it continues to evolve to meet new constraints or ideas. If it stops meeting my needs, I'll stop using it.

We use the same spreadsheet and reporting mechanism to record time spent on ongoing team-level tasks which we then use to project future levels of effort, or identify trends or that there may be a problem we need to look into.

What I've written here leaves out a lot of detail in and around what we do, including how we decide what work to do in any given case, the existence of peer review (so there are other eyes on the work), the help I get from my team in spotting bugs in what I've done, questioning why I do it - which is how this particular post came about - or suggesting improvements, the importance of trust on both sides, and the importance of testers being empowered and supported in their work.

Edit: David Hansson of Basecamp covered some similar ground in Drive Development With Budgets Not Estimates.


Popular posts from this blog

Notes on Testing Notes

Ben Dowen pinged me and others on Twitter last week , asking for "a nice concise resource to link to for a blog post - about taking good Testing notes." I didn't have one so I thought I'd write a few words on how I'm doing it at the moment for my work at Ada Health, alongside Ben. You may have read previously that I use a script to upload Markdown-based text files to Confluence . Here's the template that I start from: # Date + Title # Mission # Summary WIP! # Notes Then I fill out what I plan to do. The Mission can be as high or low level as I want it to be. Sometimes, if deeper context might be valuable I'll add a Background subsection to it. I don't fill in the Summary section until the end. It's a high-level overview of what I did, what I found, risks identified, value provided, and so on. Between the Mission and Summary I hope that a reader can see what I initially intended and what actually

Why Do They Test Software?

My friend Rachel Kibler asked me the other day "do you have a blog post about why we test software?" and I was surprised to find that, despite having touched on the topic many times, I haven't. So then I thought I'd write one. And then I thought it might be fun to crowdsource so I asked in the Association for Software Testing member's Slack, on LinkedIn , and on Twitter for reasons, one sentence each. And it was fun!  Here are the varied answers, a couple lightly edited, with thanks to everyone who contributed. Edit: I did a bit of analysis of the responses in Reasons to be Cheerful, Part 2 . --00-- Software is complicated, and the people that use it are even worse. — Andy Hird Because there is what software does, what people say it does, and what other people want it to do, and those are often not the same. — Andy Hird Because someone asked/told us to — Lee Hawkins To learn, and identify risks — Louise Perold sometimes: reducing the risk of harming people —

Enjoy Testing

  The testers at work had a lean coffee session this week. One of the questions was  "I like testing best because ..." I said that I find the combination of technical, intellectual, and social challenges endlessly enjoyable, fascinating, and stimulating. That's easy to say, and it sounds good too, but today I wondered whether my work actually reflects it. So I made a list of some of the things I did in the last working week: investigating a production problem and pairing to file an incident report finding problems in the incident reporting process feeding back in various ways to various people about the reporting process facilitating a cross-team retrospective on the Kubernetes issue that affected my team's service participating in several lengthy calibration workshops as my team merges with another trying to walk a line between presenting my perspective on things I find important and over-contributing providing feedback and advice on the process identifying a

Testing is Knowledge Work

  The Association for Software Testing is crowd-sourcing a book, Navigating the World as a Context-Driven Tester , which aims to provide responses to common questions and statements about testing from a context-driven perspective . It's being edited by Lee Hawkins who is posing questions on Twitter ,  LinkedIn ,  Slack , and the AST mailing list and then collating the replies, focusing on practice over theory. I've decided to contribute by answering briefly, and without a lot of editing or crafting, by imagining that I'm speaking to someone in software development who's acting in good faith, cares about their work and mine, but doesn't have much visibility of what testing can be. Perhaps you'd like to join me?   --00-- "We need some productivity metrics from testers" OK. I'd like to help you meet your need if I can but to do that I'll need to ask a few questions. Let's start with these: Who needs the metrics? Is there a particular pr

Risk-Based Testing Averse

  Joep Schuurkes started a thread on Twitter last week. What are the alternatives to risk-based testing? I listed a few activities that I thought we might agree were testing but not explicitly driven by a risk evaluation (with a light edit to take later discussion into account): Directed. Someone asks for something to be explored. Unthinking. Run the same scripted test cases we always do, regardless of the context. Sympathetic. Looking at something to understand it, before thinking about risks explicitly. In the thread , Stu Crook challenged these, suggesting that there must be some concern behind the activities. To Stu, the writing's on the wall for risk-based testing as a term because ... Everything is risk based, the question is, what risks are you going to optimise for? And I see this perspective but it reminds me that, as so often, there is a granularity tax in c

Agile Testing Questioned

Zenzi Ali has been running a book club on the Association for Software Testing Slack and over the last few weeks we've read Agile Testing Condensed by Janet Gregory and Lisa Crispin. Each chapter was taken as a jumping off point for one or two discussion points and I really enjoyed the opportunity to think about the questions Zenzi posed and sometimes pop a question or two back into the conversation as well. This post reproduces the questions and my answers, lightly edited for formatting. --00-- Ten principles of agile testing are given in the book. Do you think there is a foundational principle that the others must be built upon? In your experience, do you find that some of these principles are less or more important than others?  The text says they are for a team wanting to deliver the highest-quality product they can. If we can regard a motivation as a foundational principle, perhaps that could be it: each of the ten pr

The Great Post Office Scandal

  The Great Post Office Scandal by Nick Wallis is a depressing, dispiriting, and disheartening read. For anyone that cares about fairness and ethics in the relationship that business and technology has with individuals and wider society, at least. As a software tester working in the healthcare sector who has signed up to the ACM code of ethics through my membership of the Association for Software Testing I put myself firmly in that camp. Wallis does extraordinarily well to weave a compelling and readable narrative out of a years-long story with a large and constantly-changing cast and depth across subjects ranging from the intensely personal to extremely technical, and through procedure, jurisprudence, politics, and corporate governance. I won't try to summarise that story here (although Wikipedia takes a couple of stabs at it ) but I'll pull out a handful of threads that I think testers might be interested in: The unbelievable naivety which lead to Horizon (the system at th

Testing and Semantics

The other day I got tagged on a Twitter thread started by Wicked Witch of the Test about people with a background in linguistics who’ve ended up in testing. That prompted me to think about the language concepts I've found valuable in my day job, then I started listing them, and then realised how many of them I've mentioned here over the years .   This post is one of an occasional series collecting some of those thoughts.  --00-- In this series so far we've looked at words and syntax. In both cases we've found that natural language is an imprecise medium for communication. We might know the same words and grammar as others ... but they will have their own idea about what they mean ... and even where we agree there is ambguity ... and all of us, the world, and the language are evolving ... all the time. Today we'll add semantics which, in a pleasing twist, is itself ambiguo

Leaps and Boundary Objects

Brian Marick  recently launched a new podcast, Oddly Influenced . I said this about it on Twitter: Boundary Objects, the first episode of @marick's podcast, is thought-provoking and densely-packed with some lovely turns of phrase. I played it twice in a row. Very roughly, boundary objects are things or concepts that help different interest groups to collaborate by being ambiguous enough to be meaningful and motivational to all parties. Wikipedia  elaborates, somewhat formally:  [boundary objects are] both plastic enough to adapt to local needs and constraints of the several parties employing them, yet robust enough to maintain a common identity across sites ... The creation and management of boundary objects is key in developing and maintaining coherence across intersecting social worlds. The podcast talks about boundary objects in general and then applies the idea to software development specifically, casting acceptance test

Personal Development

The other day I got tagged on a Twitter conversation between a couple of my colleagues, Ben Dowen and Dan Ashby , which ended with Ben citing me as an example: But there is a trap, in that a Dev who Tests, or Tester who codes both risk becoming Test Automators ... The counter argument is Testers who code can do as @qahiccupps does, and use and build tools to explore. A jumble of thoughts tumbled out as I read it and here they are, in no particular order. It is flattering to be mentioned but I'm far from the only person doing this. Maaret Pyhäjärvi   and Rob Sabourin are vocal about the value it can bring and go out of their way to tell and teach others how to get it. Ben is right when he says I use coding as a tool, and as a tool factory. It's a means to an end. Coding itself doesn't give me a lot of pleasure. Having created a useful thing gives me an enormous amount of pleasure. I am not a great developer. But then I rarely need to be.   Yes, I have made bug fixes that