The practice of design

8 October 2024

The end of discovery research?

I remember when it was normal for me, as a designer in an organisation, to be the first person to talk to a customer about what their experience was like with said organisation. It was exciting, we were breaking new ground every project, finding out stuff that in hindsight seems obvious but at the time was revolutionary (“Can you believe, customers have added WAIT as a stage in your process?”).

The first few weeks of every project would be spent doing one-on-one interviews with customers. Diving deep into their experience, getting up to speed with what was happening, going down rabbit holes, and then spending several more weeks looking a wall of post-it notes to make sense of what we heard.

I’m starting to think those days are over, for a whole lot of reasons, some of which I’ve listed below:

It’s already been done. These days, many organisations already have bucket loads of customer research. There’ll be research reports from all the previous design projects, plus market research and surveys, a huge range of other data and statistics, and maybe even academic research into the broader field. It hasn’t necessarily helped with any given organisation getting a good grasp of their customers’ needs, and I know, you don’t trust the quality of the work that’s been done by someone else. But it still exists, and diving into a new round of discovery research feels… self-indulgent?
It pretends there’ll be a moment in time when we know what the problem is. In fact, the problem can rarely be articulated as one thing. It’s almost always a whole range of factors that are intertwined, complicated, and always evolving. An approach that assumes we can have a fixed problem statement, doesn’t lend itself to adapting either as our knowledge expands or the problem itself changes.
It separates the problem from the solution. When we’re trying to solve for something, the solution is an integral part of the problem. People’s experience is crafted by how the solution is currently designed, so if we’re trying to change that experience, we should be doing it in the context of how the solution is working and will work.
It encourages reflection rather than action. I’ve never met a designer that didn’t love research, but that can lead to a tendency to dwell. To keep digging, find out more, confirm our findings. We need to be more focused on what we’re going to change, in order to make a difference.

Fundamentally, I think we’re telling ourselves a lie that discovery can identify the problem, in a way that will let us do anything about it.

Don’t get me wrong, I still want to do exploratory research with customers. I just think we need to rethink how we go about that.

Beyond the Double Diamond

Everyone’s old favourite model, the UK Design Council’s Double Diamond, posited originally (and with reason) that we needed the discovery phase because too many projects started with a false idea of what the problem was.

This rationale for doing discovery still exists. Despite how far we’ve come down the service design path and the swathe of existing research in almost every field (if not every organisation), too often we don’t have a clearly defined problem. Or people aren’t aligned around the problem, or they’re making assumptions about the problem, or they’re wrong about what the problem is.

So, we need to think about a different approach.

Playing between the problem and the solution space

Assuming we can access the existing research, we’ll find existing hypotheses about what the problem is, and what’s working and what’s not. Start there. Instead of ignoring what’s already known, take it on board and play between the problem space and the solution space from the start.

The idea is that by putting solutions out there in a very draft format, we learn more about the problem space and can evolve solutions accordingly.

In practice this means:

We’re still researching with customers, but we’re doing it iteratively and with prototypes to test our thinking, rather than an open ended discussion guide.
We still start each session with open questions (ie. “Tell me how this currently works for you”) so customers can prioritise what’s most important to them, and identify what’s working and what’s not.
We then use some form of artefact—paper prototypes or concept cards for example—based on what we already know, to provoke a response. Remembering this is not about yes or no answers, it’s about understanding the rationale behind why something is of interest or not, to feed into our understanding of both the problem and the solution spaces.
We’ll need several rounds of customer research, with some time to reflect in between, to reiterate our understanding of the problem. Each round should involve iterating prototypes, probably creating new ones based on what we learned, with the artefacts getting more detailed as we go.

In conclusion…

This is now my preferred approach to design challenges. It’s truly iterative and keeps the focus on learning from start to finish. It’s a win-win situation. The organisation is happy because we’re building on what they already know while moving straight into action, and we’re happy because we’re getting customer input into our solutions right from the start.

18 October 2023

Why we banned powerpoint

Alexandra Almond The practice of design banning powerpoint, feedback, powerpoint

It was a very important project with a lot of senior people, and everyone was still figuring out what they were doing when we got involved. There was a lot of busywork going on, including two people who were pulling together a powerpoint pack of the history behind the project. At first glance that seemed like a reasonable thing to do, but two weeks later they were still working on it. They didn’t know what it would be used for, had no clear audience outside the project team, and were spending most of their time fussing about layout and fonts.

We estimated based on their salaries that the pack had cost $40,000 by the time we intervened. So in an attempt to get focus back onto the things that mattered, we banned everyone from powerpoint for a week.

It was a shocking move, and we completely underestimated the impact on people. To the point that someone called in sick the next day, and the team appointed a delegation to talk to us about it. The fundamental issue was, these senior leaders couldn’t understand how to do their job if they weren’t creating a pack.

That is an extreme example, but most organisations are guilty of focusing at some level on the output and the communication more than the work that needs doing.

Why is powerpoint such a problem?

The first challenge is that powerpoint starts with the form, not the function or need. Your thinking is constrained by the confines of the report; the headings you’ve included and what you can fit on a page. It gives you the illusion that you’re making sense of things because you’re ordering your thoughts, without helping you to actually think through those things in the way a synthesis process would.

Secondly, it makes it look like it’s finished, which sends a message that further input is not required or desired. It’s easy to create a powerpoint that looks good, but that encourages you to go straight from “thinking” to “finished” without doing the feedback bit in the middle. And that bit — engaging people around your thinking — is far more important than what it looks like at the end.

Why do we like to create polished, finished things?

It’s complicated, and we probably all have different reasons. For the public servants I mentioned earlier, creating the deck was how they framed their job. They knew what they had to do and there was a sense of security and satisfaction in doing it (even if it was the wrong thing).

For designers it’s also about identity. We want what we do to look beautiful, because it represents who we are and the work we’ve done. We think if it looks messy, people will assume the work is messy. But in fact, presenting a work in progress in a polished way is not going to make you look good. You won’t have all of the answers, so why make it look like you do?

A possible way forward

I was recently introduced to the concept of the 10–50–99 approach to feedback and we’ve since implemented this successfully at one organisation we’ve worked with. The premise is that you both present different materials and ask for different types of feedback, depending on how far through the project you are. So at:

0% you might not have anything to support what you’re doing, maybe it’s a conversation about the purpose of the work to see if you’re aligned.
10% it’s hand drawn or scribbled notes at best. Maybe you’re even getting people to create something with you. Feedback is conceptual.
50% you might have a rough mockup. Feedback is still quite conceptual, checking we’re heading in the right direction.
99% it’s pretty polished, and the feedback you’re looking for is spelling and grammar.

I would ban powerpoint at 0, 10, 50, and only use it at 99. It’s easy to put stuff in a powerpoint at 50% but sends the wrong message and won’t give you the response you need.

Having said that, I’d love never to use powerpoint again. I’ve absolutely worked on projects where there has been no polished deliverable at the end, and it’s great. They’re almost without fail impactful projects because the focus is not on the output, but the outcome of the work.

13 June 2023

If personas are the answer, what the hell was the question?

Alexandra Almond The practice of design archetypes, design, mindsets, personas, stereotypes

I don’t like personas. They’re human-washing; like greenwashing only for people. A way for organisations to pretend they care about their customers without doing the work to understand them.

Why do they exist?

Like many wrong things, the intent behind personas isn’t bad; to help build empathy and understanding of the people you’re designing for. And for those who do the research to create them it probably does do that job, but for anybody picking up the finished article and trying to use that to design something, it’s almost sure to go wrong.

Naturally, bureaucracies love personas. They’re such a nice, clean way of bringing structure to something that’s otherwise very messy; i.e. humans.

But the thing is, humans are messy. We say one thing and do another. We change our minds and our opinions on a pretty regular basis. We have unreasonable expectations and we’re biassed in ways we don’t even realise. We’re irrational. And no two of us are the same.

Persona of a designer

Meet Luke. He’s a designer, can you tell? It’s a bit of a giggle, because I know at least a dozen designers who could be Luke. Hey, I fit several of those criteria myself.

From a marketing perspective these things might be useful to know. But nothing on that page helps me design something that will work better for Luke. For example the owner of a coffee shop might look at this and decide to combine the most common 3-element drinks so instead of ordering a decaf soy latte, you just order a “Designer” coffee. Luke might love this, but he might also never go there again because he feels judged.

And there you have the key issue with personas; they’re stereotypes. They don’t encourage you to know and understand people, they encourage you to judge, and make decisions based on what you think you know about those people.

That’s fine when we’re making fun of designers, who are well paid and in relative positions of power and privilege. It’s much less fine when you’re doing it to people who aren’t. Then it’s just ick.

Take people with accessible needs

In all the research we’ve done in the transport space, and all the people we’ve spoken to who are in a wheelchair, EVERY SINGLE ONE IS DIFFERENT. They have different likes and dislikes and backgrounds and opinions. How they experience the world and navigate public space is different, and nuanced. The fact they’re in a wheelchair is just one element of their lives.

There are commonalities that you can design for, of course, but understanding the nuances is what helps you design more inclusive spaces. And you can’t possibly understand that from a conglomerate version of wheelchair users pulled together in a single page with some jaunty comments and a cute image.

When personas go bad, they’re harmful

True story. Someone walks past the project space, stops to read a persona on the wall and says “wow, this one sounds like a dick”.

Judgement. Not empathy.

Obviously the way the personas are described can play into that. The worst one I’ve seen had a woman who would “walk across the platform and swing her hair and strut in her high heels and look around to see who was watching.” Maybe they were trying to bring the character to life, so you felt something for them. But you can’t possibly understand the people you’re designing for with a caricature.

Personas can emphasise the differences that people have, which might be great, but not if it’s used to profile or misrepresent people (which it too often is). And if there are things people disagree with it can drive a negative response, leading to a greater divide rather than building empathy.

So what can we do instead?

A better way to build empathy is stories of real people, who are willing to share this way. At Meld we’ve done that really effectively with videos of different people navigating spaces, for example. That works as a communication tool and an empathy building tool.

We’ve tried mindsets and archetypes, with mixed success. The mindsets we developed for public transport recognise that the same person may make vastly different travel decisions today and tomorrow depending on the task at hand, who they’re travelling with, the weather or any number of other factors. This at least brings the why in — traditionally personas are all what — but it doesn’t help much in terms of design.

There’s no shortcut for getting human input into your design work. You need to do the hard yards and try things out and iterate them when you see what works. You need to watch your users — as many and as varied as possible — interact with what you’re designing and find out why they behave the way they do and get them to help you make it better. It doesn’t have to be onerous (see my previous post about five ways to prototype, for example) but it does have to be done.

Please. Let’s not put any more personas out into the world.

10 May 2023

One way to make meetings more effective

Alexandra Almond The practice of design commit meetings, effective meetings, input meetings, intentional meetings, meetings inform, meetings reflect

Why are we doing this? What is the outcome we want to achieve? If you’re clear on this, your choices are much more obvious and decisions easier.

Applying this to meetings

A practical example is meetings. At Meld Studios we recently did some work with Who Gives A Crap where lots of people struggled with meeting overload. This is not uncommon in organisations, as I’m sure you’re aware. So we started exploring ways we could reduce the number of meetings.

There are lots of ways you can have fewer meetings. Send an email instead; use asynchronous feedback tools; schedule meeting free days; ban recurring meetings; give people more power to make decisions so they can just go away and do whatever it was they were going to meet about.

That last point in particular made us think we might be headed in the wrong direction. Because in many cases, meetings aren’t taking you away from work, they ARE the work. They get a bad wrap, but done well meetings can be how we collaborate and get input and also build trust with each other.

No, our actual intent wasn’t to have less meetings, but to make meetings work better.

Intentional meetings

So then we went meta, and applied the “clear on intent” mantra to meetings. We asked people to be intentional about setting the scene for the sort of meeting they were having, and what the expectations were of everybody in that meeting.

We identified four meeting types, with the type informing the style of the meeting:

Inform is listening. It’s largely one-way communication. Whoever’s presenting has things to say and they’ll do it as efficiently as possible.
Input is where you’re expected to contribute. This is more like a workshop, where the meeting owner has prepared things they want participants to feedback on or help them develop further.
Commit is about making a decision. Participants should be up to speed with the background before coming in, and the focus of the meeting is not going over old ground, but agreeing a way forward.
Reflect is using reflective practice. It’s more collaborative and less agenda driven, with participants coming together to share reflections and experiences, with the intention to learn, improve and grow.

How’s it working out?

It’s going pretty well, I think. Having a shared language is a shortcut to a common understanding. And it’s generally agreed it makes the meetings themselves quite pointed and more effective.

Meeting invitations start with the one word, before the description, so the last invitation I got said “REFLECT: Meld & WGAC catch up”. I knew exactly what we were going to be doing in that meeting. It would be easier if the options were embedded in the technology (and yes, they’re looking at that) but it’s not hard to just do it.

One side effect of having to be clear on the type of meeting was it made some people realise they maybe didn’t need a meeting. So maybe we did end up with less meetings, even if it wasn’t our ultimate goal.

Feel free to adopt this intentional meetings approach for yourself. I’d love to hear how it works out for you.

22 December 2022

Die, deliverables, die

Alexandra Almond The practice of design clear intention, deliverables, expectations, reports

The longest report I’ve created was 120+ pages of text heavy detail that was probably never read outside the project room.

The only purpose of reports like these is to make the client feel like they’re getting value for money. They don’t help the organisation move on or change anything, and despite the quibbling over phrasing during the approval process that might almost make you think this shit really matters, most of them will get put in a drawer and forgotten about.

Can you tell, I don’t like reports. I don’t even like the good ones that are really succinct and insightful and visually appealing. Unless I can see what the report is for — what purpose it has beyond simply documenting the project — you can count me out.

The same thing goes for any deliverable that is created for the sake of ticking a box based on what the client thinks they need. Journey maps, personas, blueprints, storyboards. They can all be useful but if you’re creating them because they were in the brief, and not because there’s a pressing and urgent need for them, then they’re a waste of everyone’s time.

Oh I know sometimes you’ve got to bite the bullet and make an artefact. But the most impactful projects I’ve been involved in haven’t had any kind of “thing” as a deliverable. My favourite is an organisation we worked with recently where the deliverables were 13 mini pilots in progress, and not a report or journey map in sight.

Fundamentally we need to rethink what the outputs of our work as designers could look like. Here’s some ways you can do that.

Set expectations up front.

Be clear that you’ll create USEFUL things not just things the client thinks they want because they saw someone else had one and it was cool. That means you probably won’t know at the start what the deliverables are going to look like; that will emerge through the process.

I’m aware not everyone’s going to be okay with that uncertainty. Pick your clients, and use stories of how things ended up different-good to sell the possibilities. And if you have to commit up front to a journey map and a set of personas (ugh), sow the seed in your proposal and your kick-off session that maybe we’ll find something that will work even better.

This works best with clients who already have a design mindset, or clients who know how you work. Repeat clients make this easy and fun.

Be clear on intent.

Be really clear about WHY you’re creating whatever deliverable it is. Make sure you know why people want it, how it will be used, and the outcome you expect from putting this thing in the world. You’re then halfway to creating something useful.

For example, we had a client who wanted a toolkit and a training program to address their capability gap. Of course it turned out that the problem wasn’t capability; it was culture. That’s an over-simplification, there were a lot of different problems. But none of them were things that could be addressed by a toolkit and a training program. Once we’d identified what the actual problems were, we could start exploring what things might address them.

Explore your options.

Don’t just jump to the obvious solution or thing that you could create, instead explore your options. Take into account the intent and think about how best you can achieve that; maybe it could be a visual, a checklist, a pilot, a process flow, a spreadsheet, a conversation, a video, a board game, a way of working, or something else.

Test your deliverables the same way you’d test your ideas during the project. If you’re thinking about creating a video, make sure you try the concept out with your target audience and find out where and when they watch videos (if they even do), what length will be most effective, the style or format that works for them, and so on.

If all else fails…

After all this if you really, really have to write a report, apply the same principles. Make sure you’re clear on the intent, tailor it to the audience, and make it as useful and practical as possible.

2 August 2021

Synthesis: it doesn’t matter how you categorise your post-it notes

Alexandra Almond The practice of design categorisation, Connection analysis, Edward de Bono’s Six Thinking Hats, sorting data, Synthesis

The archetypal image of my industry is a designer looking pensive in front of a wall full of post-it notes.

Sorting post-it notes is easy enough, but that’s analysis, not synthesis¹. When you’re doing synthesis, your goal isn’t categorisation. It’s to understand not just how this data fits, but what this data means. And that meaning is often completely unrelated to the categories you’ve put the data into.

You do still need to sort your data

So if it doesn’t matter how you categorise your post-it notes, why do we bother? We bother because the act of looking at each data point and making a decision to put it somewhere, gets your brain working and making the connections you need to make sense of everything.

I have a colleague who believes you could sort the data into alphabetical order and it would have the same impact. It’s the act of sorting, not what you sort it into, that’s important. That might be taking it too far, but there’s definitely something there. Have a look at how Melis Senova talks about using your subconscious and intuition in sensemaking. Jon Kolko also talks about the sense of magic in the synthesis process. As designers, we’re largely agreed that the value of synthesis is not in the sorting of data, but in something else that happens either as you’re doing the sorting, or afterwards.

More often than not the insights I end up with are completely unrelated to the categories my data is sorted into. This freaks out new designers particularly, who like to be able to trace back where something came from, as if that demonstrates its worth. Clients often like to see these linkages as well.

Yes, you need to be able to justify your insight, but not necessarily by following a linear trail from x people said this so y must be true. You get to the “So what? by prompting your brain to work in different ways. I’m reminded of Edward de Bono’s Six Thinking Hats; sorting your data is the white hat for the facts, but you also need to bring in the red hat for intuition, the green hat for possibilities, and the yellow hat for optimism. It’s not a perfect metaphor but it’s a useful way to think about synthesis beyond just data.

Some ways to get the magic happening

The challenge with teaching people synthesis is that everyone’s brain works differently, so what works for me won’t necessarily work for you. My advice is to try out as many things as you can and see what prompts those magic “aha” moments for you.

Here’s some ways you might like to go about it:

Write down your random ideas: Capture those fleeting thoughts as they happen and put them somewhere to come back to later. This often takes the form of questions for me. “Why do people love blue so much?” or “Imagery=Emotion but not for C21?” or “False hope?!!!” These are actual post it notes from a recent project. Many of them won’t go anywhere. Others prompt conversation and ideas. You won’t know which thought is useful until much later.
Walk someone through it: Externalising my thinking is what works best for me. I talk through what I’m seeing with someone, and that helps me to understand it. If I could only do one thing it would be this, and I’d do it every day.
Pull out the key anecdotes: There’s always particular stories from research that you find yourself telling again and again. Pull them out and examine them. Why are they resonating? How do others react when you tell them?
Plot it on a journey: Yes, journey mapping. But in an exploratory way rather than for communication purposes. You could plot activity, or look at emotional state over time, or just compare the best to the worst journey.
Opposing force analysis: Identify tensions where two things may be true at the same time. “People don’t want to be guided” and “People have a better experience if they’re guided.” They’re both true. Identify those moments and then ask yourself, so what?
Connection analysis: Look for connections between themes. Sometimes they jump out at you, sometimes you need to force yourself to see them. Try putting two seemingly unrelated things together and see what happens. Do these things mean something different together than they do in isolation?
Visualisation: Try drawing what you’re thinking. Sometimes our synthesis wall is full of little vignettes, calling out what we think is happening.
Challenge: This is best done with someone else. Challenge each other about what you’re seeing or thinking or writing down. Is this really true? Is it representative? Why is it important?

To close…

It’s not an exact science, and it’s something you have to learn by doing, but you will get better at it every time you try. Happy synthing!

¹I don’t want to get hung up on terminology, but for the sake of this article I use the terms synthesis and sensemaking interchangeably, to talk about the thing that happens in the “define” stage of the double diamond. And yes I know that’s not how everyone thinks.

1 October 2020

5 ways to use prototypes to learn, not test

Alexandra Almond The practice of design blank space, create a prototype, prototype scenarios, prototypes, sorting cards, walkthroughs

Let’s start with a definition. For me, a prototype is a thing you use to provoke a response, so you can learn.

In many fields and even traditionally in the design world, prototypes are used primarily to test whether your designs are working. That’s great, and we do that too. But using low-fidelity prototypes much earlier in the design process, lets you learn heaps more and different stuff than you would in a traditional depth interview.

Here’s some examples of different types of prototypes we’ve used to learn, rather than to test.

1. Scenarios

This is a very recent project, looking at ways we might work flexibly post COVID-19. We created some deliberately unrealistic future scenarios, and used them to prompt a reaction. The point wasn’t to find out which of these scenarios was the best, it was to understand how people were thinking about work.

We could have just asked what they thought the future of work looked like. But this way instead of a whole lot of “middle ground” scenarios, we got their raw reactions to the extremes, and used that to frame with them what possible futures might look like.

2. Make your own prototype

Getting people to build something themselves, reveals what’s important to them in a way they might not tell you (or might not be able to articulate) if you asked.

We used this in the design of a new service centre. We gave participants all the elements that needed to be included, and got them to choose where each element should go and how much space it should have. They also had options to add their own elements.

This worked particularly well in small group sessions, as a way to promote conversation about the relative importance of different elements, why they are important, and how they relate to each other.

3. Physical space walkthroughs

We used low-fidelity prototyping in an existing physical space, when we were helping the State Library of Victoria define their future state vision.

It wasn’t just about the space, it also involved people role playing different scenarios in the space. The scenarios were deliberately loose so participants got to improvise what they would do. Observing this, we learned what might work in practice and some of the likely challenges.

4. Card sort

This one isn’t revolutionary, but it’s really useful and we use it all the time.

You develop a range of concepts loosely connected to your research area, and get participants to sort them into categories — “I don’t think so” — “Good idea” — “Game changer”. The idea is not to identify which idea is great, but to understand why and how people think about the topic you’re exploring.

5. Blank space

This is one of my favourite recent examples, because it’s so simple and effective. This is testing that we’d normally do on site, but had to do virtually because COVID.

Before we started the testing, we put this image up on screen, and asked participants what they’d expect to see, or like to see, in that blank space. Their answer tells you so much about them — their familiarity with the location, how much notice they take of the thing you’re talking to them about, what they’re most interested in at that moment.

So there you have it. Five simple ways to put prototypes in front of people early and easily, that will help you learn more about them and about your subject matter.

25 February 2020

Make policy great again

Alexandra Almond The practice of design government policy, policy, policy in business

Most policy practitioners agree in principle what makes for great policy advice. The Australian Government has developed a model (read more here on the Policy Hub) and has articulated great advice as being:

Clear on intent
Well informed
Practical to implement
Influential

The Australian Government model for great policy advice.

This isn’t new — we’ve known for years what great looks like, but we’re still struggling to get there. The myriad of scathing reviews and assessments on public policy implementations in Australia are testament to that.

The thing is, doing policy consistently well is going to take change — big, scary, long-term cultural change. And when you’re stuck in the day to day grind of “this is the way we’ve always done things”, changing anything is difficult, much less changing the world.

It’s okay, there’s still hope! We’ve been working with federal, state and local governments and know lots of small things you can do to have an impact. Here are three to start.

1. Ask why

One common policy failing is not being clear on what problem we’re trying to solve. That’s because we make assumptions, or just do what we’re directed to do rather than seek clarity.

I get it, you don’t want to look like a fool by asking someone up the chain a stupid question. I’d like to say there are no stupid questions but in a bureaucracy that’s not how it plays out. So find a safe way to do it, maybe test out your questions with people around you to see if you’re making sense, then take a deep breath and call someone who might have the answers.

Every senior person I know says they’d rather you ask a polite clarifying question up front, than waste time working on the wrong thing.

2. Find someone who has a different opinion to you and listen to them

You’re living in a bubble. We all are. Push your boundaries and find someone who isn’t in your bubble and listen to them.

Don’t assume you know where they’re coming from. Go and spend time with them and try to understand who they are and why they think that way. Challenge yourself not to defend or rationalise or try to change their mind. Just listen and nod and keep asking “Why?” and say “Tell me more about that” whenever they run out of steam.

They’re unlikely to change your mind, and you’re not allowed to try and change theirs in this scenario, but that’s not the point. You’ll be surprised how much broader and better your thinking and your policy advice will be when you bring in different perspectives.

3. Test your thinking with someone, before you make it pretty

In government and in business, we’re rewarded by getting things right. You spent hours on that powerpoint presentation and you need to make damned sure it’s perfect and you can answer any question that gets thrown at you.

That’s great, but if you created that presentation in isolation, it’s not going to be right and it probably won’t be good. And you’re unlikely to get constructive feedback because everyone will be too afraid of hurting your feelings after you spent so much time on it.

Instead, try sharing what you want to say BEFORE you spend time crafting the message. There’s lots of ways you can do this but my favourite is to draw what I’m thinking, and then use that as a prompt for a discussion. Like the diagram below. This drawing helped convey what I was thinking, and nobody had problems giving honest feedback on something this rough. The fact I put it out there like this showed I was genuinely open to feedback.

Rough and ready sketches generate better feedback.

So there you have three things you can do today to improve policy outcomes.