September 8, 2021
Building Customer Success for Scale with GitLab and Sanity.io
In this special fireside chat, GitLab’s Head of WW TAMs and Sanity.io's CEO explore the work of founding and scaling an enterprise focused...
In episode 7 of Unintended Consequences, Yoz and Kim continue their conversation with civic design expert and author Cyd Harrell. They unpack the systemic goals and features of private tech and government, and expound on ways the two could coexist more effectively.
About the Guests
Cyd Harrell is a civic design consultant, well known in industry and government for her creative approach to UX research and service design. She is the author of A Civic Technologist’s Practice Guide.
Yoz Grahame: What are the most common kinds of failure that you see in projects?
Obviously, this kind of giant, inflated procurement problem is one.
But even the well-intentioned ones that seem to be keeping their costs down, where do you see these things hitting the rocks often?
Cyd Harrell: A big one, one that's dear to me is they haven't been able to do any research.
And so it doesn't actually work.
That is a huge and common case that is tragically avoidable.
And so I think that's interesting because it doesn't match up to our ideals of scale.
Typically, research is done one-on-one or one-on-small group, and then used to draw inferences about what the larger group needs.
Which may or may not then be supported when you get to a point where you can measure something at a bigger scale with analytics, or a technique like that.
People don't agree on what it should be, actually.
So a new administration comes in and goodnight. That happens all the time.
And it's a real caution on the timescale piece to work with the less political parts of government a lot of the time.
Yoz: Yeah. Seen a bunch of good projects get killed that way.
But actually, let's go back to the research for a moment, just to clarify, because people might not be aware of the kinds of research you're talking about here and what that looks like in practice.
Could you give us a brief snippet of what that kind of research is?
Cyd: Right. So my core discipline is user research.
Meaning, mostly actual observational research of people trying to use some technology product.
And what that tends to do is give you the whys and wherefores of what is wrong with your product.
So analytics can really identify the whens and what was happening at the time that the problem happened.
But it can't tell you anything about the motivations or the context in which somebody is using it.
I think some of the trickiest, having documents that download in Word.
Guess what? Many people don't have Word these days.
And fillable PDFs as a stepping stone to online forms, very, very hard to do on a mobile phone.
And if you go and meet people who are waiting in line to fill out those forms and see them try to do it on a phone, you'll easily see, can't do this.
This doesn't make any sense.
Often, actually the piece of benefits eligibility that I think is really sticky, usability wise, is how people report their financial situation.
So I know the Code for America group working on GetCalFresh has worked on this a lot.
When people have income from gig work, for delivery driving or something, they don't have a sort of, "Hey, look, put your last pay stub and tell us how much you make every month."
It's give us an estimated profit and loss.
And so helping people ask those questions in a way that people will actually give the right answers, especially on things where it's under penalty of perjury that you have to give your answers, it's a really good idea.
Kim Harrison: Yeah, that's terrifying. I would just choose to not report.
That would really be a deterrent for me to participate in whatever that thing is.
Yoz: Yeah, the threat of legal action.
Cyd: Huge deterrent, but that's one of the policy pieces.
Kim: Gotcha. Yeah.
Cyd: Like, well, we want people to take this really seriously, so we're going to have them swear.
Kim: Mm-hmm (affirmative).
Cyd: But then you engender this terror of getting something wrong, especially if you're in a situation where it's not easy to just say, "Yeah, my income is the thing on my last paycheck."
Definitely have a number of things like that in the court space where I often work.
The paperwork in court cases is really scary, even if we're talking about the civil justice system, which is mostly where I work.
But for people who switch off and go, "Oh, the criminal justice system is more exciting," civil justice includes everything about family law, from child custody to divorce, to domestic violence restraining orders.
It includes eviction. It includes guardianship and bringing a suit against a neighbor in a small claims court.
All those things are civil justice and you are not guaranteed representation in any of those kinds of cases.
And so there are really tricky usability things with the way to ask questions.
And in a lot of cases, it's part of the law or part of the rules of court that they have to be asked in certain terms.
And those are not necessarily the terms that are clear to somebody who doesn't have a lawyer, or even somebody who isn't a native English speaker, or is English-speaking at a reading level.
Yoz: It ties into actually, one of my favorite aspects of civic tech, to do with opening up different kinds of contributor roles, which is writing.
So many people think, "I can't do civic tech. I can't contribute because I don't code."
And this is like, "No, that's not the most difficult language."
Programming languages are not the most difficult languages we deal with in civic tech.
Cyd: By the way, I couldn't say because I don't code myself.
Yoz: I find code so much easier to wrangle, mainly because it's unambiguous.
If there's ambiguity, then there's a flaw in the programming language. That's a bug.
Whereas writing English that you want to be sure that people understand, just English, I'm not even dealing with any of the other languages.
And so the role of content designer as it's now called, if you want to go into that a bit more.
Cyd: I don't know of a civic tech team that's strong that doesn't have content designers as a key part of it.
And that is probably not true of every team in the private sector.
You tend to see it more in marketing teams or various specific sectors.
But communicating, just between the government experts in a particular policy or service and the public users who are often not experts in the policy or service, is really, really hard.
And we spend an enormous amount of time on plain language, better language, language that actually works for people.
Yoz: It's vital. I imagine one of the frustrations with research is talking to people who have misunderstood what laws apply to them, what their situation is.
Cyd: Yes. That can be heartbreaking.
And I worked on this study last year, where people were trying to get unemployment benefits and we were trying to report to policy makers.
And one of the biggest disappointments that people experience is when they thought that there was some situation in which their society had their back, and it doesn't.
So in the context of the pandemic, everybody thought that they would be able to get unemployment.
It's an obvious situation where government steps in to help.
And when talking about America's small government movement, I often say there is some situation in which you believe the government should assist you, even if it's pointing the right way out of town when a storm is coming in.
And sometimes, some people have a really hard time imagining themself in a situation where they would need more common types of assistance from the government.
But really people do, nearly all, have a fundamental belief that as a society, we should help each other out.
And when the pandemic appeared, one of the I think, most consensus points where we would all say, "Government help is needed here for many, many people," and they couldn't get it, it was intensely painful, whether that was a first experience or a re-experience, to see that help not arrive.
The help that did arrive was very, very welcome.
It did an enormous amount for the people who are able to get it for the amount of time that they had it before things started changing, to get people forced to come back to work in some cases before it was safe to do so, and having to make difficult choices because of that.
Or benefits ran out. When it comes down to it, Kim was asking at the beginning of, why would you do this?
Cyd: This is a reason why.
If we can help fulfill that promise or make it more real by using some of these skills we've come by, one way or another, it's enormously appealing.
Yoz: Also, highlights.
Certainly what was so disappointing to me about the pandemic response in America, and even Britain, which mostly had a terrible pandemic got this better, was messaging.
A complete lack of coordinated messaging that should have been driven from the top-down, through state and city level, and could have helped so much.
Cyd: It's so funny.
I think several times during the pandemic, I've exchanged messages with our colleague, Cordelia Yu, about what the messaging ought to have looked like.
At Thanksgiving last year, the guys from Queer Eye should have been in masks, touring the White House with the First Lady, showing how she was creating a distanced celebration that they were going to do on Zoom.
And that is just communication at scale failure, which is a whole other category of a really interesting thing.
Yoz: But that tends to be the kind of failure that we often see the egotistical tech types come in and say, "Oh, we can solve that. We should come up with a way to solve that. There's an app. We'll come up with something that can solve this."
That is way beyond the typical comprehension of what happens at this scale.
Cyd: Yep. One of the really interesting cases maybe to think about is Nextdoor, which was intended for neighborhoods to communicate with each other.
And grew into something that really perpetuated racism in the US through this encouragement of people to report things that they thought were out of order.
And encouragement of white people in particular, to better safe than sorry, report things. Call police, do all of this.
And they have taken some steps to address it.
But if you want to consider, it's a public civic sphere thing with a private company, with the government not involved, but calling on government resources to oppress certain people supported by this technology that was theoretically intended to do something else.
That's a real nest of what we're talking about today.
Kim: Yeah, exactly. Because government is used to thinking about my constituents are everybody, not just teenagers that like to post photos. But it's everybody.
Kim: It's for the entire community, like, "I love Instagram."
Some people are blind. It's not exactly the funnest thing for them.
There are things that are fun.
But yes, if this is stepping in to a space that maybe is more for government, like government is used to thinking about, who is this for?
Cyd: Yeah. Well, we hope, and is doing a better job thinking about it.
Yoz: I loved how you talked about there are places where everybody expects that government will serve them, that government will have their back.
And so much of what government does is between the cracks, is between the lines of what people think government does.
You have this mental image of all the ways that government works.
And most people don't realize the vast majority of government interaction is not in those groups. It's not in those topics.
Cyd: Right. And the idea of government efficiency is really interesting, with this idea that government can step in to support us in any type of crisis.
Well, it's a lot harder if the government is at a bare bones staffing level, because we don't want to pay taxes to support government at a resilient level of staffing. With backups.
Yoz: The topics that I would love to discuss further, or for you to go into further, there's a quote here.
One of your quotes actually, from the book is, "I focus on principles and sets of questions to help technologists find the right way to do the most good, starting with finding the people already doing the work."
That was one of the major problems that we see in civic tech, is that people who are eager, turn up and jump in, imagining that nobody else is there already.
And quite often, there are loads of people in there for ages.
Cyd: Yes. And often, if the shiny tech people look shiny and seem exciting and innovative, they get way more attention than the people who have been working for years.
It's a huge problem.
Yoz: If you could give advice to people listening, who are either interested in civic tech or making private commercial tech better, what advice would you give them?
Find the people doing the work. That is a single piece of advice. And don't offer that you are doing something cool and new and you want their help. Offer to help them. Find out what they need.
And it's vanishingly rare that there isn't somebody doing the work.
They may not be a technologist.
They may not see themselves as working in technology.
But there is almost certainly somebody working on it too, and nearly, always somebody with a similar idea.
And sometimes it's really interesting to find that people who had the idea of a few years ago and tried to make it go, and it wouldn't.
What happened? And what are the conditions that have changed that make you think you might be more successful now?
But just generally taking a perspective that not that you are a super smart person because you know technology, but that you have a cool and valuable skill that you can put at the service of somebody's mission that you align with.
Which I think is a very different posture from, we are the smart people who can come in and solve things.
Yoz: Absolutely. The humility in that makes a huge difference, especially when dealing with people who have been in government for ages.
The thing I often think about that is common amongst all the civic tech movements I've seen is they found the civil servant who's been there for 20 years, who knows the lay of the land, and most importantly knows which threats are the important ones.
Cyd: Yes. And convinced that person that they were worth investing time and political capital in.
Cyd: Usually, by showing humility and showing capability, and exciting them about possibilities.
I love meeting those people.
They have so many amazing stories, and I learned so much from my public servant partners.
Kim: Talking about partners, Heidi had a question.
What would make regulation more of a partner than a constraint?
Because people think government, they think regulation.
And it's difficult to get anything done.
And what are these heavy things that are constantly getting in the way of me needing to operate at the pace that I think is appropriate?
Cyd: Right. So regulation sits midway in the stack of government rules.
You probably have a statute that underlies a regulation, and the regulation probably has interpretations and guidance for people that lie on top of it.
So figuring out what animal you're actually dealing with.
If it's guidance, then the local secretary or head of agency or whatever, might have the ability to change it.
If it's regulation, it might take more than that.
If it's statute, then you have to work with a letter.
And we're once again, out of the realm of technology and into political advocacy and lobbying, which is sometimes a negative word, but talking to members of staffs to get a rule change.
The only way I know to work well with compliance and regulation people is to accept that all of this is well-intentioned and has a good purpose.
And then with that in mind, to be able to show why it is or isn't serving that purpose in this particular way, and ask the people who know how it works and how to build it, what options we have for how to make that change.
Or there are techniques from technology, like building a prototype, that shows how something could be different.
That can be really useful in our process, but it isn't a technical process.
Cyd: If you think about an agile and multidisciplinary team, what it might look like in the private sector, in government, it should almost always involve compliance from the beginning.
Maybe they don't come to every standup, but maybe you check in with them once a week.
Kim: Mm-hmm (affirmative).
Cyd: We're thinking of doing something that involves this regulation.
What should we watch out for?
What are you absolutely sure we can't do?
Where might there be a little wiggle room?
And you have to build a trust with that person to be able to ask those questions, but you don't want to have them as a final reviewer who says, "Wait a second, you can't release this."
Kim: That touches on something else, because we did, we pulled together a list of questions; digital transformation and continuous improvement.
So it sounds like making sure that you've got all the right people in the room helps you move faster, rather than like you said, waiting til the last second to review this and realize that it's--
Cyd: Yeah. And you know what? It's interesting that you put it that way.
Because right, you might think, "Oh, well, if we move fast in the beginning," just like get to shipping, sometimes that can be the right strategy.
But it isn't always. And so figuring out where you need to move fast and where you need to bring in more people to move more together and more strongly, there's a cliche.
If you want to go fast, go alone. If you want to go far, go together.
Kim: Mm-hmm (affirmative).
Cyd: A lot of government stuff has to go together.
I talk about two stages a lot.
So this was showing what's possible, the thing that we do.
There are people who have never experienced launching a website in less than three years.
And if a team of civic technologists can show them that actually, we can put something together in a couple of months, in fact, we can change it on the fly while we're in a meeting with you, that can be a really important moment to help them shift toward a mindset that we can do things more nimbly.
Kim: Mm-hmm (affirmative).
Cyd: But we couldn't probably build a production grade website for something that included really important personal information at the scale of a country, in a couple of months.
So then there's this whole doing what's necessary stage.
Comes after showing what's possible, to make sure that we work through the infrastructure and the security and the privacy.
And I think that technologists have to earn it a little bit to work into those areas of practice and those areas of the stack, to be able to help do something not just at scale, but life impacting at scale.
Kim: It's such a different pace of doing things, because when I go to work every day and I think about the things that I have to do, it is a very different case.
It's a very different way of operating. Neither is wrong. Neither is correct.
Cyd: Right. They're suited to different situations.
Kim: Yeah. And ultimately, what's my goal? Who am I trying to help?
What am I trying to accomplish? And remembering what that is.
Cyd: Very much so.
Kim: What is one thing that you would want to leave people with, that you would want people to think about government a little differently?
Cyd: Government is as good as we make it.
And it's a quote from, I believe Barney Frank, that, "Government is what we do together."
And the lesson that I hope we all take out of the pandemic is that we need to be strong in a way that happens when we do things together.
And we should spend time and resources and yes, money on a government that is first-class.
And technology is one of the things that we can use to build a government that serves us all and is inclusive, and makes us all feel like we are really part of a robust civic sphere. But it's not the only thing. And so anything that you do that furthers that, that robust civic sphere that's for all of us, that helps.
Kim: Tell me a little bit more about your book.
Because I would like to include, what inspired you to write this book?
Cyd: Civic tech in this current incarnation, there have been people trying to improve government forever, but in this current incarnation, it dates back to around 2008, 2009.
And there's been increasingly growing participants and some successes.
But if you look at the changes that people really want to create through this work, you're looking at things that are going to take decades.
And so we need to put a flora under the field and say, "Here's where we are. Here are some things we've learned that we can just package nicely on a little book, and you don't have to learn them through pain and hard work anymore."
You can just hopefully, read this nice, little package and get some of them.
And then we can all move forward to the next set of mistakes and things to learn.
So my goal in writing it was really to make everybody doing the work feel seen, and also provide that object that somebody can hand to somebody who's interested in the field, to give them the basics.
Kim: And what do you see for that next generation?
If you were to do a V2 of this book in another 10 years, what do you hope to see happen between now and that next moment of, this is where civic tech has taken us?
Cyd: In 10 years, I hope that very many governments will have public service minded technologists as part of them.
And that people will be working with technology that has non-commercial goals and socially good goals, a lot more as a legitimate path in the technology industry, too.
Kim: Mm-hmm (affirmative).
Cyd: And so, one of the things I really hope is that it will be a lot easier for people to come out and do this in the early parts of their career.
It's often the paths are really hard right now, and there's a ton of coaching work that in some ways, is suited to people who have a few years under their belt.
And I hope in 10 years, you'll be able to say, "Yeah, I want to be one of the technologists who support all of our benefit systems."
And that, that will be both something that you have a path to and something that looks great on your resume."
Kim: So interesting. Yeah.
Because right now, I can't imagine like, you know how to do a thing and you could do it in a private company, or you could do it in this space.
But that isn't something that I think about.
Kim: The space all on its own.
Cyd: I hope it actually won't need that much updating in 10 years, because it mostly isn't about specific technologies.
So I hope it either needs very little updating or is almost obsolete, because the field is more established.
And one of the things I keep saying is part of the reason I wrote it is there really wasn't a shelf of books, and I wanted a shelf of books.
And so every time I talked to a group of civic technologists, I asked them if they have a book in them, because we need that shelf of books.
And we need institutions.
And we need places that people can go to find out about the great work that's happened before, the field work that's happened before.
We need those resources of a real field.
Kim: In our notes here, somebody mentioned it and it's making me think.
We're talking a lot about the people who are responsible for the technology.
But I feel like there are other skillsets necessary, other people necessary to manage this technology, think about how it's utilized, who maybe don't come from a technical background, who are going to be hand-in-hand.
And how are those people involved?
And how should they be thinking about it in the future, so that it's just a natural given that they're considering these tools that have become-
One of the recommendations that we made in a strategy project, I was originally on the California Department of Technology, for the executive branch was certain technology basics should be a part of management training, just like certain financial basics are now.
It's a realm where anybody who's managing teams and managing spending on these kinds of things, needs to have a basic understanding.
Kim: Mm-hmm (affirmative).
Cyd: And one thing we didn't even get into is how government outsources a lot of technology now.
And so it feels like this separate realm.
And seeing it, or whatever succeeds in our current world of websites and apps based on databases, be just part of the work stream of providing services, or part of the work stream of implementing policies and regulations, without this extra outsourcing and gap between the people who really understand the policy and the people who work on the tech, that is a future that we really want.
Kim: Mm-hmm (affirmative).
Cyd: And your colleague is right about, I can't call myself technical.
And interestingly, civic tech has a really high percentage of designers and content strategists, as Yoz was saying.
People from non-technical technology disciplines maybe, is a good way to put it.
Kim: Yeah, for sure.
Is there anything else you'd like to add to this?
Having this perspective is something that we don't get very often.
You're not the kind of individual we get to talk to and think about, or see the perspective of very often in the space that we're in.
Cyd: Interesting. See, I would like to see those things come more together.
Kim: Yeah. I think it's because we are such an early stage startup.
Maybe in a couple of years, we'd be the kind of company where we're used to talking with a broader range of customers that are interacting with their use cases and how they operate.
But where we are right now, we're very much at the cutting edge corner of where we're focused with early adopters who are able to run quickly and be risky in ways.
But I think organizations like government, not quite yet.
We need to be sure short because we have an impact on people that really matter.
Cyd: Yeah. There's a place for government taking a less protective risk posture on some things, but move fast and break things.
If you break the unemployment system, it's not so good.
Kim: Back in the days when we could show up to conferences in real life, I remember being at a conference where people were talking about how public groups, government groups were using existing technologies to reach their constituents, but how they also need to be careful about what they were doing is depending on what the platform was.
It could put those people at risk.
And so for instance, using Facebook as a tool to reach students? Wonderful.
But how much of that sensitive information is secure and not being used in negative ways?
Cyd: Everything that's happened with Zoom when school has gone online?
Cyd: That's been a really interesting set of questions.
I don't think there were a lot of better options available.
And on the other hand, it's not ideal in a number of respects.
Kim: Yeah. And so, what do you do?
Because there are these teachers that need to reach their students.
You've got people who want to do things.
But how do you help those people who may not have a technical background, better understand the platforms they're trying to leverage?
So they can use free or very inexpensive things to accomplish the goal without causing issues. And oh, it's complicated.
Cyd: Yeah. But some of those building blocks, like Yoz was talking about with the US web design system, making those basic capabilities more available to the public sector, I think will make an underrated significant difference.
For remote interactions becomes part of the public sector toolkit, that changes access to courtrooms.
That changes access to schools. That changes access to all kinds of things.
And we still have to be careful of people who just can't get to that access, but it's potentially a leap forward.