Tag: weeknotes

Focus and freedom

Team focus for this week has been about figuring out how to take away work they shouldn’t be doing so they have the space to do the work they should. They struggle with focus, feeling that they are responsible for everything that happens with their product and I need to help them become leaders who guide the business. The two areas of work I’m going to tackle first are support queries from customers and delivery management. These are things that are somewhat within our team’s control so they are quicker wins for freeing up time for the PM’s.

I had a good chat with one of the team about shifting roles to be more of a Delivery Manager. She said yes, as long as I could tell her what to do and when to do it. I replied that we could certainly agree on what she should be trying to achieve but that I would never tell her exactly what to do as I think we should be shaping roles to fit people, not the other way round. I want her to be able to use her strengths to achieve things rather than feeling like she’ll always be struggling against ways of working that don’t recognise her as an individual. Communicating can be done in lots of ways. If your a people-person (whatever that means) you might do it by talking to people, and if your an accurate-detail-oriented person then you might do it by providing information in a spreadsheet. Neither is wrong and both can achieve the same thing.

Next week I’m putting together a training budget to get my team to conferences and on training courses. This is part our ambition to show the Product team as experts and leaders in the business rather than order-takers for features. I still want to see if I can arrange a visit to the HMRC Product Team, and I’ve also been thinking about how we might be able to get knowledge and insight from the BSI Sector Teams in order to help us understand the wider market of Standards. I also want us to arrange customer visits soon. I know it’s a scary concept for some of the PM’s, and they’ve hinted at this with comments like, “but my customers are internal people who request changes to the product”. No, those are your stakeholders, your customers are the people that use your product to achieve something they value, and you need to understand what they want and why.

Who is responsible for the solution?

I spent some time thinking about how and why work is sized. There seems to be a weird organisational thing where if a piece of work is sized as small or medium it is considered not complex enough to need a Business Analyst to work on it, but if it’s sized large or extra large then it’s regarded as complex enough to need the support of a BA. The really weird thing is that Product has to pay IT for using their BA’s, even though IT do the sizing and we don’t get to decide whether a BA is required. This led me to thinking that if we sliced all of our work small enough it would never need BA support, we’d save money and deliver more more quickly. But that led me to uncovering the gap in how products are built and maintained which I think explains why it’s so hard to get anything done right; there is no solution design.

Solution design is implied, I think, but it’s never explicit. For small and medium work the assumption is that the solution design has previously been done before the work request was submitted. And for large and extra large work the assumption is that it will be done during the business analysis process, although again, those assumptions are never made explicit and the work of designing the solution to achieve the outcomes for the customer and not adversely affect current business is never done.

We have work request documents, high level business requirements documents and user acceptance documents, all of which describe the end state (often in different ways) but we have nothing to describe what is going to be built. Without a shared understanding of what to build and how it is going to impact other business processes, it’s no wonder things get delayed, built wrong, and increase the burden on UAT to reveal all those misunderstandings. The question I’m left with is who is responsible for solution design? It fits into the Development phase of our product management process, which is in the solution (rather than the problem) space, but I’m concerned that the PM’s will be the only ones with enough cross-functional nous to do it effectively, but really they should be focused in the problem space.

Using the process to design the process

Had a really good workshop with two of the team to decide how we want to set up Aha to reflect and support the new practices we’re going to be using. It was interesting to see the understanding of the how we want to change our practices grow during that two hours but most importantly we made decisions. That seems like it’s often a difficult thing to do. That needs to change. Product Managers might not be responsible for making decisions but they should be responsible for the pace of decision making.

As for making the change, we’re getting clearer about what it’ll look like. We’re using the Discover, Define, Develop, Deliver process discover, define, develop and deliver our Discover, Define, Develop, Deliver process. Part of that is about how a piece of work moves through those steps in Aha from being an Idea in the Discovery phase, being promoted to a Feature for the Definition phase, the Feature being prioritised on the Backlog for the Development phase, and then added to a Release for the Deliver phase. The other part is showing the team the tools, techniques, outputs and metrics from using the process and using Aha. There are lots of tightly held assumptions that ‘this is how we work’, that we’re going to unravel and challenge the PM’s to shift their focus from the development and delivery of features to the discovery and definition of opportunities. They are going to need to think bigger, and I’m going to need to put the things in place that let them do that.

When do you do your homework?

We’ve been doing user acceptance testing on the latest round of work. It has been haphazard at best, precisely the opposite of what good testing should be. We test based on our understanding of what we were expecting the results to be because of our familiarity and knowledge. Not a good way to test. All of the PM’s left their testing until the last day. This is behaviour that tells me that a) they don’t prioritise testing over other work, and b) they don’t really want to be testing. It’s like leaving your homework till Sunday night.

It’s another area of work I think we should take away from the PM’s but its going to be a bigger shift than removing support queries and delivery management, so it’s going to take a lot longer. And it’s not just a case of telling the QA team that they are now solely responsible for testing as the quality isn’t there yet. So, part of the shift needs to be building up the QA team to be able to test with more rigor and contextual understanding, and the even bigger part is nailing the solution design so that testing can be against the impact of the work not just does it technically do as expected.

Tweet of the week

My favourite tweet of the week was from Ben Holliday, Chief Design Officer at FutureGov. I’m a big believer in hustle and creative problem solving over blindly following procedures. I know the PM’s in my team do work this way sometimes, but it’s definitely a skill I want them to improve upon. I’d rather they were figuring things out by talking to people, scrawling on postit notes and trying new things than sitting on their own typing things into a spreadsheet.

It’s my second week at BSI and I feel like my rate of learning has increased since last week and I’m getting a clearer picture of how things work and what things need to change. I don’t think my thinking is quite big enough yet but I’m happy to focus on fixing foundational things over the next few months.

The conceptualization I developed last week to describe the three areas of work is holding true. People, Process, and Product are still my priorities and I’m getting a clearer idea of what needs to be done with each

Process: learning by doing

I’ve been increasing my understanding of how the current product development and management process should and actually does work.

There are lots of smaller parts to the process that need drastic improvement, such as how we prioritise work for development without 3 and a half working hours spent force ranking a list of features into priorities based on knowledge only one person holds. We need to think about how to formalise the prioritisation as ‘seizing opportunities and avoiding risks’ so that ultimately the work can be perpetually prioritised in Aha and the IT Project Managers can access it any time they want, but that’s some time away yet.

But before we get into those smaller optimizations we need to get the overall process in place. We going to experiment our way towards the best process but we’ll start by using the double diamond to give us four broad phases: Discovery, Definition, Development and Delivery. To make it clearer for everyone I think we need to map the diamonds to functionality in Aha, some tools to use, and the expected outputs.

Discover

This is the ‘why’ of the entire process. It’s where we should ask bigger questions about whether this is an idea worth exploring, whether it seems to fit with our strategy, and what outcome we think we might be able to achieve for the customer.

We use ‘Ideas’ in Aha for this, which is where we record the outputs from our discovery work which might include market research, customer interviews, internal infrastructure investigations, etc. This stage is meant to be expansion and all about collecting everything we need to understand the problem but we shouldn’t be coming up with solutions yet. Once we’ve done enough discovery work we’re score the idea on it’s desirability, viability, feasibility and strategic alignment to decide whether the progress with the idea.

Define

This gives us the ‘what’ that we’ll be working on. It’s where we take the expansiveness of the ideas that we’ve agreed to progress and tighten it down until we converge on a solution. Some to the tools for doing this might be a cross-functional team workshop and writing user stories.

In Aha the Idea gets promoted to a Feature and this is where we’ll continue to record all the work we do to ensure that we don’t lose any useful information or insight. When we’re ready, we’ll push the user stories from Aha into Azure DevOps for the IT team to see what they’ll be working on (although they’ll already know as they were involved in the discovery stage).

Develop

This section gives us the ‘how’. It’s the part of the product process where the developers work on the user stories that we wrote in the Define stage.

In Aha the Feature will be prioritised in the Backlog, which will replace the need for prioritisation meetings with IT Project Managers as they’ll be able to see a perpetually prioritised list of things to work on.

Deliver

And this part is the ‘when’. It tells us when the work will be released into the production environment.

In Aha the Feature is added to a Release that matches the release schedule that the IT team will be working to.

Discovery and Definition cover the problem space and is where we should be spending most of our time, and Development and Delivery are in the solution space.

In the medium term we’ll working towards turning ideas into opportunities rather than thinking about each idea leading to a single feature in isolation. Then, opportunities should lead to outcomes for customers, so for example an idea to change the content delivery network would be part of the opportunity to improve the infrastructure of our products, which adds to the customer’s outcome of using secure, reliable and scalable products.

Experimenting with our processes

Next week we’re going to start our first experiment.

• Why: Inspect & adapt to improve. Safe-to-fail space to try different things. Evolve towards a process that is easier for us, more fit-for-purpose and achieves for the business, and delivers more value to our customers faster.
• How: Pick a problem area within our process, understand the problem (discovery phase), agree what change we’d like to happen and how we’re going to measure it (definition phase), think of lots of ways to solve the problem and pick one to try (develop phase), make the change we decided upon and see what effect it has (deliver phase). So, we’ll use a product development process to experiment with and improve our product development process whilst learning how to evolve the process.
• When: Every two weeks, starting 17th June. We’ll try meeting on Monday morning to pick the problem and experiment with the solution over the following two weeks.
• What: We’ll spend a bit if time choosing our problem-to-solve for our first experiment. I’d suggest that we start with parts of the process that are within our control, e.g. creating a shared knowledge bank by working in the open and keeping Aha Feature cards up to date with the progress on that Feature so that over time.

Challenging myself to not go rogue

In previous roles ownership/accountability so I could go off on my own to get things done. I have to keep holding myself in check in this role to prevent myself from doing that. Now I need to take a more responsible for direction/leading role and temper bringing people with me with understanding where they want to go and helping them get there.

Working in the open is definitely not part of the culture. I even had comments about my wall (a rogue behaviour I’m not keen to lose) being wiped clean because of the risk I might share business secrets. That tells me it’s a problem-to-solve and one we’re going to work on as a team.

People: changing the space the Product Managers work in

I’ve been looking into which stage in our current process for developing a feature the PM’s spend most of their time, and it looks like it’s at the development end where they seem to fall into multiple roles of doing a bit of admin work, bit of project management, bit of business analysis work, bit of delivery management, and then spending a bit more time fire-fighting issues and liaising between IT to make sure work is delivered. This really isn’t where they should be spending their time, but I think I’m beginning to understand some of the reasons why. There’s a bit of the PM’s not having a clear perception of what their work should be, and a bit of a lack of trust in others. Whereas other organisations might have delivery managers to oversee that part, here the product managers feel like it’s their responsibility.

There are two parts to changing that; one around removing that work from them to give them the space to do the things they should be doing, and another around changing their aspirations and skills so they know what to do and how to do it.

I’ve started looking into the production support process, which isn’t very clearly defined, to try to remove the first line support from the PM’s role (and perhaps moving it to a Customer Success role in the future). And I need to spend some time with the Product and Project Managers to try to help them get clearer ideas around how is responsible for what during the development phase.

I had an idea that if I could find a similar organisation with a more mature product function perhaps I could arrange a visit so we can see how their product managers work. I thought of a few different sectors to look at but like the idea of visiting the HMRC product team. There seems to be just enough similarities in that both organisations provide information services to businesses without being too similar that we fail into the trap of trying to copy what they do. We somewhere aspirational not instructional. So next week I need to work on making it happen.

I have to keep reminding myself that process can change far more quickly than people are likely to so I have to keep sense-checking myself that all these things are on the right trajectories so that they can meet at the right point in the not to distant future.

More broadly than within the team, I spent a bit of time thinking about the stakeholders for the products I’m involved in and what their motivations are. It’s clear (if a little frustrating) that most people just want to do their job without having any interest in their industry or sector, and are motivated by an external locus of control. This makes it a little easier to understand their hopes and fears in the short term but it makes it much more difficult to encourage improvement over the longer term. People that live and breathe their industry generally want to get better at doing what they do. Anyway, understanding people’s motivations behind the decisions they make is something I want to develop over time.

Product: breaking the cycle

I started a new piece of work and progress a couple of others. I’m conscious of the need to keep the pipeline of work full because that is the expectation at the moment, but I haven’t yet got much of an understanding of the capacity of the pipeline or how we can optimise it. I also learned a bit about how the financing works and have some ideas about how we can change it to be no longer on a feature by feature basis but fund a set run rate for development work and then optimise our pipeline to match.

Some of the feedback I’ve heard is that the right people aren’t always involved at the right stages of developing a feature. The idea of ‘starting together’ seems like a good experiment to try in to improve on this. Every time we want to do something we get all the relevant stakeholders together (either in person or virtually) and we Brain Dump (copyright Suky Semhbi) everything we can think that might affect the work we’re considering. It can become one of our Discovery phase ‘tools’. I’d hope that doing this kind of thing won’t only align the right people early on but that it will also build better relationships between the teams.

The bigger challenge I’ve picked up on this week is a chicken & egg situation of where a team can’t invest in providing data services for products without a clear justification and benefit, and the product team can’t build things that deliver value without the data services from the other team. I think the way forward is for both teams to step back, stop thinking of it on a feature level and come up with more direction for data and products together.