Transcript
Speaker: Welcome back to the PolicyViz Podcast. As always, I'm your host, Jon Schwabisch. On this week's episode of the show, I'm joined by David Waters, CEO and founder of Simple and Engaging, a firm that helps nonprofits with their technology solutions. It's a really interesting conversation. I know you're ah you're thinking, well, I'm a data viz person. I don't need to listen to this about you know operational problems and people having issues with their ah HR and their executives, but...
Speaker: we talk in this conversation about organizations having especially non-profit organizations having challenges with their organizational frictions right and so one person has their core data work but then they've got this little other solution over here going on and they have to solve this thing in this spreadsheet so they got to go talk to you know sam over there and mary over there and so we talk about the work that david does to try to help non-profits streamline, improve those processes, be more efficient and be more productive. And so if you are working in an organization, particularly an organization that is working with data, this conversation I think can be really helpful to you as you try to think about how you might improve your own work processes, but also the work processes and goals and workflows of your team or of your entire organization. So this is a little bit different than maybe some of the previous episodes you've been listening to because we're not talking specifically about data visualization or data communication, but really talking about data workflows, which, of course, if you're trying to get stuff out the door, is also really important to that entire process. So here's my conversation with David Waters only this week on the PolicyViz podcast.
Speaker: David, great to meet you. Welcome to the show. Thanks, John. i Appreciate it. Glad to be here. I know people probably heard your voice for the first second and I'm going to say this and be like, I feel like I should say hello in Spanish, but I think people probably listening or like, wait, why would you be talking in Spanish? So I I do want to understand sort of the, where the company is located and why you are in Mexico city. So maybe we'll just get to like your background, you know, the company's background, and then we can, we can get into all the good gory detail stuff.
Speaker: Yeah, yeah, no, that sounds great. So um yeah, again, thank you for having me. Delighted to be here. A little bit about me first, and then and then I'll ah explain the the Spanish reference. So i've I've spent most of my career sitting in the the gap, I suppose, between technology and the people that are actually trying to get work done in various different guises and different forms over the years. And what really interested me early on was and I saw this far too often, was was seeing organizations spend massive amounts of money on technology that you know didn't really solve the problem that they had or in a lot of cases, and and quite ironically, actually made their lives harder. And it's that kind of situation that led me to start
Speaker: ah simply in engaging. the The business was founded and run out of Australia. Today, a lot of our work is with smaller organizations and ah nonprofits. and And we look, broadly speaking, where work is harder than it should be.
Speaker: So, you know, things like repetitive processes, reporting that's really slow and clunky, information that's getting stuck between different systems, um and and more recently, you know large amounts of data that that people don't really have the time or capabilities to to make sense of. And our job really is working with those organizations is to help them to to work out if in some cases, but how technology, automation, and technology
Speaker: you know ah like a lot of people, more increasingly AI can actually help that that that problem. um We work with folks in in the UK, North America, Australia, New Zealand, which means that the you know my my time zones are a bit all over the place. But um we've actually relocated my wife and I about four years ago to Mexico City. Time zone-wise, it works really good for for North America, for the UK and and for Australia. um And we you know we also love the idea of our, we've got a four-year-old daughter now. We love the idea of her growing up bilingual.
Speaker: um it's ah It's an absolute gift. um So she learns nothing else from yeah from the two of us as parents, we've we've at least ah we've at least done that. So so yeah, we're're we're very happy. It's a very cool city. yeah and And yeah, it works out really well business-wise.
Speaker: Yeah, that's great. I mean, i've ah I've only been there once, had a lovely time, just great city, great food. So that's amazing. um I want to ask, so you started with talking about small nonprofits. What, generally speaking, like what are the size of the nonprofits that you're working with?
Speaker: Because I've worked with nonprofits at three people or six people, and that's like really small. But like what's the size range? and And let me ask the follow-up question, which is, is it generally your experience that there's like a data person in these in these organizations that's not really a data person, but they did something cool or they did like a thing and now they're like thrust into this role?
Speaker: Yeah, so we've we've worked with nonprofits of all sizes from, you know, one full-time person and a couple of volunteers or, yeah you know, in some cases, a group of volunteers all the way up to, you know, probably where...
Speaker: it starts to make a little less sense for the nonprofit is sort of about 150, 200 people mark. And really the the main thing that we see that that kind of changes that is how they view technology and what sort of resources they have available within a technology group. So, you know, up to that sort of 100, 150, 200 mark, they may have a technology department, but it's usually more geared to servicing the organization, you know,
Speaker: you know, does making sure Microsoft Office is working as it should do, making sure that just the the kind of back end stuff gets done. And and typically they they don't have the resources to take a more strategic view um of what that looks like. And and also in in those smaller organizations, you know, I think what's really important in the work that we do is making sure that it's not just a technology lens that we're looking at this. You know, the the goal of any piece of work that we do is to make work easier to run.
Speaker: um so you know sometimes that involves a lot of technology sometimes it involves you know just some better processes and and better ways of working and and and things like that so um you know the type of folks that we'll be working with they may have people that take on some of those responsibilities or have been given those responsibilities but it's not their core skill set um it's something they've inherited it's something that you know, they've become in some cases the kind of go to person to to do that rightly or wrongly. It's probably not even in their job description, you know, most of the time. So it's ah yeah, it's really interesting how that kind of dynamic shapes and and as organizations get larger, it's it's there's I think you're
Speaker: you're starting to see that there's people who are already trying to look at what that looks like. We've we've turned down work before in larger organizations as well. I think um for me personally, I work, I prefer working more closely with, with organizations to the people that have actually got their, their boots on the ground, as it were. um You know, it's, I feel very fortunate in the work I do. I get to work with very smart, passionate,
Speaker: hardworking people who are just trying to make the world a little bit of a better place in their different areas, right? And to to play a part in that is great. So um so yeah, that they're kind of the the groups that we look to work with.
Speaker: Gotcha. And do you, it sounds like you set them up with a plan to acquire the right technology package, whether it's software or hardware or AWS or whatever it might be. Do you also help train them or do you point them in the direction of we're going to get you into Salesforce or Airtable or whatever it is. And here are the resources where you would go to train up your whole staff.
Speaker: Yeah, so we started life doing more of just the pure consultancy piece. So like, hey, you're looking for... you know, a new project management software. Let's have a look at how you work. Let's have a look at different options. And here's one's a different price ranges.
Speaker: um A lot of the times, what one of the the options that we'd come up with wouldn't necessarily be a a a single platform that was off the shelf. It might be a series of platforms that you kind of stitch together, which, um you know, achieve what you want to do at a much lower price point.
Speaker: um folks would come to us after that. you know That's how we started life. And they're like, this sounds great, David. We don't have the skill sets in house to be able to develop this out.
Speaker: you know, broadly speaking, could you do it? So we shifted that kind of process to be more of a managed services operator as well as that kind of consultancy piece. So, you know, we take a very, you know, very kind of strict view of this. we're not We're not a reseller of specific platforms. We're not tied to, you know, other other organizations that do what we do. They're, you know, Microsoft Office Houses or or whatever.
Speaker: you know, certainly a place in the world for them. For us, it's more about how can you how can you understand what's really causing the the kind of friction in the work and how can we deploy technology that's going to to help that? And in you know the fun case, I geek out about this, the the bit that you know I really like, it's not necessarily deploying technology across the whole process. It might just be looking at where where it gets really messy, really time consuming, really frustrating, pulling that bit out and just automating that bit.
Speaker: you know So really taking a bit more of ah a pragmatic view on this rather than how do we turn this into a massive technology transformation program, which um you know historically, unfortunately,
Speaker: doesn't have very high success rates. so Right, right. So you've already mentioned this word friction and that and the phrase operational friction appears on your on your website a bunch. um So maybe you could help people understand what that means, but also, I guess you've already kind of defined it. So I guess I'm also curious, like what are the biggest frictions that you're seeing in these, you know, let's say one to 200 person groups.
Speaker: It is an interesting sector to be in. And I would suspect that the frictions that these sorts of firms are having are obviously going different than a place of thousands of people, but also similar, you know, similar size organizations in different sectors. But yeah, so so what does that sort of like operational friction look like? And what are the places where you see it the most frequently? Yeah.
Speaker: Yeah, absolutely. So the the term operational friction in in the context that that I use i is really a bit of an umbrella term for all of the little things that make work harder than it needs to be.
Speaker: And if you look at them in isolation, a lot of them look pretty harmless, right? You know, so it's someone downloading a file and, you know, an Excel file from a system and cleaning up every month. It's, ah you know, and you see this all the time, you know, someone who's got their own little side spreadsheet because they don't quite trust what's in the platform that they're supposed to be using. yeah um You know, it's, it's a,
Speaker: a monthly report that takes three or four days to to produce um or the ones that we the one that we see quite often in in nonprofits, especially in smaller organizations, is that that kind of go to person. So you you hear people saying stuff like, ah Sarah knows how to do that. you know Go and ask her.
Speaker: And individually, none of these things are ah are a massive problem on their own. But the problem is, is that a lot of the organizations that we look at have dozens or hundreds of these all together.
Speaker: Right. The bit that I think is a testament to people is that it's the leaders and the team, they've all become really good ah working around all those workarounds. Yeah, yeah, yeah.
Speaker: They're having to remember the things that for whatever reason the system doesn't or the process doesn't. they' They're you know copying information all around. there They're creating little fixes here. you know They're working...
Speaker: they're working you know, extra hours, they're staying late, they're working weekends to make sure stuff gets finished. And, you know, these folks are ah really acting as is what I started to call as like, almost like human shock absorbers.
Speaker: You know, they're they're absorbing that gap between how the work is meant to happen and how it actually happens. And we we did some research this year and and and found this in in in the research that we did. We looked at two years worth of of nonprofit um discussions on Reddit.
Speaker: um So it's about 5,700 posts. We found that people were far more likely to ask, how do I deal with this? How do I take this on? Rather than...
Speaker: Does anyone else work like this? Why are we doing this this way? How can we think about redoing it this way? And that really came up in the data when we we we did really detailed sort of qualitative analysis on it. And for every 19 times a constraint was mentioned in a post, only once was there some sort of mention in that post about challenging it.
Speaker: So, you know, that that sort of stat really stuck with me, because you know, sort of 19 to one, because you've got, you know, This situation where work keeps getting done, clearly, you know people are working really hard, they're pushing really hard, but the underlying problem is actually getting worse.
Speaker: And the kind of the interesting sort of dichotomy on this is from a leadership perspective, things might look okay. They're seeing the work getting done. They're seeing the output, right? We need that funder report. Okay, we'll get that done. But underneath you've got folks who, and in a lot of time in smaller organizations, it's also the leadership in this situation. um You've got folks who are you know like ducks on water, right? Desperately paddling like crazy to stay afloat. And and that has a cost.
Speaker: And, yes you know, an organization either pays for that in, you know, lost time, limited capacity, um frustration a lot of the time. And, you know, increasingly,
Speaker: In nonprofits, it's very well documented. Burnout is a massive problem. Right. You know, so we've we've got this situation now that that we're seeing in organizations where, you know, operational friction, that big umbrella term, survives because you've got these good people um who have got very good at working around it.
Speaker: Yeah. Yeah. So I call you, I run my firm of my nonprofit of whatever, a couple dozen people. And we have the situation you just described. I'm doing a bunch of things over here and Sam is doing a bunch of stuff over there and Mary's doing much stuff over there. And everybody's sort of like, whatever, we've all got our little side spreadsheets or whatever. That's the perfect example.
Speaker: When I call you up and say, we need help, how do you identify like, what is the thing you're going to fix first? Yeah. Like that seems a big, it seems like a big challenge to figure out like, yeah.
Speaker: Yeah, it's ah and and that's usually so so to preface this conversation, it's usually reason why people don't want to do this sort of stuff, because it does feel too big. And also for a lot of folks, they're just desperately trying to keep their head above water.
Speaker: You know, they don't have a lot of time and space to to look up above the parapet. um You know, normally I'll look at um I'll look at three things. you know So when you're looking across an organization, you're um I'm looking at three things. you know How often does something happen?
Speaker: What's the the kind of level of of you know how much time it takes up or how annoying is it? And you know importantly, what does it stop you doing? So you you know frequency, friction, consequence, right? um you know So you might have a...
Speaker: a 15 minute work around once a year, probably not something worth worrying about, like an annual rep report and they you know people have to do this work. But if you've got something that takes 15 minutes you know and there's a work around that eight people are doing every day, yeah you know now you've got a problem, right? Like now you've got something that's that's you know that that's that's that's quite you know sort of quantifiable. and Yes, the time is an issue, of course, but that time always, you know, the time that involves always has consequences, right? So, you know, maybe it delays a funder report, which means might put some of your funding at risk. Maybe it's, um we see this in a lot of cases, you've got a program lead that's spending two days doing admin instead of working with partners and deepening relationships and, and,
Speaker: you know improving services. um you know Maybe there's one person that, you know like we mentioned before, the the Sarah example of you know everything's got to go through. um you know so So you look at those three things.
Speaker: You can also look at things that kind of cluster together as you're doing that analysis. right so This also happens quite often. You've got people, they might be complaining about five different things, right? But they all involve the same process, the same spreadsheet, the same system or the same, you know, handoff between two different systems or different departments or groups or whatever. um So you probably don't actually have five problems.
Speaker: You've got one problem showing up in five different places. Right. You know, so that gives you a good kind of cue on what to look for. um And finally, another area that that I like to to kind of poke and prod around is starting with something that people really dislike doing. um You know, so, you know, sitting down with leadership teams, getting leadership teams to sit down with their team and and ask them, you know, what's...
Speaker: what's one part of your job your job that you you just dread? Every time it comes around, you're oh God, you know, it's, yeah you know, and you'll get a quick answer, right? Because people people know that, they've got a really visceral answer to that. um And maybe it's the monthly report, maybe it's,
Speaker: you know having to copy information from one place to another, or it's constantly trying to chase higher ups down for for approval. And you know these are really good kind of first problems, because if you can make one of those noticeably easier,
Speaker: people feel the benefit straight away you've you've alleviated a pain point there straight away and you know you've you've created an evangelist for looking at this sort of work already and we see this a lot in the work that we do we'll go in and we'll fix one small problem right And then we'll be asked to do more and more and more because you've got other folks in the organization saying, well, hang on, if you like if you can fix that, can like I hate doing this. Can you do this? um You know, and you've you know, and it's not just about how can we, you know, save people time necessarily to squeeze more out of them. This could be the difference between someone having to work weekends or not.
Speaker: Or it could be the difference between someone being bogged down with admin work and actually doing the job that they were hired to do and that they had the bit of work that they really loved doing. you know so in my experience, that that you know small, you know visible win really helps kind of generate demand for change across the organization far more effectively than you know going, okay, we're going to do a transformation project and here it is and everyone's going to change.
Speaker: you know That kind of top-down approach. so Yeah, yeah, yeah. So I'm curious on two parts of this because you talk a lot about you talk a lot about the people.
Speaker: yeah like Like this conversation, you have focused really on the people and not the technology. And I want to, yeah ah there's two things I want to talk about there. um the The first one is, well, I guess I want to start with the way you've described how you do the work.
Speaker: is sounds very data driven. yeah But I'm curious how you think about collecting the data because I could imagine it would be really easy, right? If you came into a into a firm and everybody said, here's the estimate of the extra time it takes me to do this task. And then you would just, you know, by task and you just sum them all up and you would say, okay, so what are the biggest returns on investment by number of of minutes spent on this thing? yep um so So what is your data collection process like when you're working with the client? When they're like, you know, so you just said like, we have this little thing, but it's what bothers me the most. So let's fix that. Let's get the small one and that'll branch out. But how do you measure that sort of return to know where to focus your attention?
Speaker: Yeah, it's a great question. So you've got Typically organizations will will come to us, they'll fall in one of three buckets. One is they've got some funding to do something new or expand on something. So it's a growth problem. Like we we need to expand this. We've we've hit ah a limit. So that kind of helps focus our attention.
Speaker: um You've got folks who are, we're doing well as an organization. but we're just drowning. Like we, you know, our our best, there's stuff we're now not doing because, you know, just the, the, the, the sheer, you know, kind of admin burden has crept up on us. And then the third will be something's gone wrong.
Speaker: So we're supposed to do this thing and it didn't, and we need to, so one in three, you know, that helps direct our attention for that second group. You know, it's, it's, I actually spend more time working with with leadership teams first to understand what's the, you know, for a better term, the opportunity cost.
Speaker: So what is it that you do want to be doing? What are, you know, you've you've hired these people, you've hired smart people, you've got a good team who don't have the time to do the things that you want to do. What are they not doing to understand what that looks like? And, you know, it might be, you know, looking at how they can improve programs. It might be how they can,
Speaker: you know drive better drive better outcomes or better relationships or whatever that looks like. And then try and understand, well, what's the time involved in doing that? So we've got an idea of what we're shooting for.
Speaker: We need to say, we need to free up 10 hours of people's time. then you can kind start going to those you know teams. and And again, we look to start small because you know budgets are always tight. you know ah There is a healthy skepticism of you know deploying technology like this and and solutions like this. So we start small, we start visible you know to build that sort of momentum behind that.
Speaker: But it's really looking to kind of coordinate what do you as a leadership team and it might be the leaders themselves or their teams what do you want to be doing that you're not doing what time do we need to free up to do that and then looking at okay well for the smart people that you know can do those things where are they spending their time you know a good activity I ask folks to do is is spend a week doing this um you know and you don't it's literally on a bit of paper by your desk just write down some of the stuff that you're spending time on and you know
Speaker: having a think about how much of it is repetitive, how much of it, you know, enjoyment comes into it as well. That's always a good way to, as I said before, to find that ah stuff. um Yeah. And really sort of build out what what that looks like. But it's it's all about trying to build that around what are the outcomes that folks are trying to achieve.
Speaker: Obviously, you're talking to a lot of people at these at these organizations. is it a combination of one-on-one discussions, kind of like focus groups, small group? And then do you, I guess I'm sure this depends on on one of those three cases, but like, do you also try to find time to talk to people people who work at the organization separately from the executive level. Yeah. Because i I totally agree with you. Like if you put someone in front of the president or CEO of their organization and could have an honest discussion about like, what are the pain points? People would have the list right away. But I could also imagine people are like, I'm not going to say anything. I don't want to make anybody mad. I don't want to get fired. So like, you try to separate that a little bit to get that like real honest in-depth feedback?
Speaker: Yeah, definitely. And and it's it's really important. And, and you know, the the way we frame the conversation with both the leadership teams and the folks actually doing the work is around we're looking to understand how the work gets done, not looking to assign blame or or, you know, identify faults in what's happening. Because, you know, a lot of the times, you know,
Speaker: will be in situations where if the leadership team there hasn't kind of prepped the conversation well enough with us, there's some defensiveness there.
Speaker: So folks are looking at that going, okay, well, I know what you do for a living and you're looking all about, you know, how we can drive efficiency, technology, automation. I'm a bit nervous in my job now. And, you know, it's ah it's that's why it's so important to come from the top down. This isn't about this is about we want you to be doing this good stuff, the stuff that we hired you for, the stuff that you enjoy, the stuff that doesn't drive you crazy. and To do that, we need to reduce this stuff.
Speaker: So that's why that that kind of framing is really important. But you've got to speak to the people that actually do the work because there's such a big gap. And we see this all the time between how people think work gets done. and how that work actually gets done. Yeah, yeah, yeah. And it's like, oh, and then I go into this spreadsheet and I check this code here. And then I'll actually have to check with Brian on this one because, you know, Brian knows what that field actually means. And, you know, then I put it into here. I don't really know how to do this bit. So I send it to, you know, Jerry in marketing and Jerry helps me out. And what people think is a straightforward process to generate a report or whatever, yeah you know, you find ends up, you know, going about 16 different ways before it ends to the end.
Speaker: Yeah, yeah, yeah. Okay, so so my second part of the people technology question. So we've been talking a lot about people. um And you you also just mentioned like, you know, just have a piece of paper on your desk and just write down the things that are you know taking time. And I'm curious if an example or two where a firm has come to you maybe with their instinct or their initial sort of discussion with you is like, we need a tool, we need a technology to fix this thing. And after your discussion, you're like, no, no, no, that's not what you need. You need this other more human capital hiring, firing, whatever it is. Like, like, when did you, an example where you talked a firm out of a technology solution and more to a people solution.
Speaker: Yeah, yeah. um So we yeah we had an organization and they were referred to us by by someone else. they'd Someone in their leadership team had been at a conference and um they were super excited about this presentation they saw on Agentic AI.
Speaker: And when they spoke to me, they were like, look, we're all in. Let's turn up the dial. Let's get agents. Let's spin them all up. Super excited about it. And it was great to sit. It's genuinely great to see people really engaging with technology in that way. And I understand the kind of mentality behind it, right?
Speaker: a lot, there's a bit of, actually not a bit of, I think quite a lot of like almost guilt around AI at the moment. yeah You know, you've got leadership teams who who know AI is important. They know it's here. They, you know, all they're seeing all the hype and they're really concerned about falling behind.
Speaker: So there's this pressure to be able to turn around and say, oh, yeah, you know, we're using AI, you know, tell the board or their stakeholders, you know, something more than just turning co-pilot on for their team. right Yeah. um So, you know, there's all in on agents and, we're you know, they're getting really, you know, sort of detailed into that. And, you know, through that conversation, I just kind of asked them to take a step back and said, look, where where are you struggling? Where is your team actually struggling?
Speaker: And you know through the process that I mentioned before, we you know we found a bunch of different things and you know there were a couple of areas. you know One in particular, it was this repetitive process around i you know website content and and social media um that they just the person had just had enough of and it was wasting so much of their time. now you know could you Could you build an agentic AI solution for that?
Speaker: Absolutely, of course. But the reality, once you have a look at it is massive overkill. Massive overkill, huge costs. What they actually really needed for that piece of work was pretty bog standard, straightforward automation that was available you know four or five years ago.
Speaker: yeah um And now, because there's you know a lot of you know sort of more user-friendly platforms to do this sort of stuff, it's work they could probably do themselves. And and that was the advice we ended up giving them. So we we actually didn't work with them in the end, you know through that process. And yeah you know my accountant doesn't like me for this, but it's kind of to to you know turning work down. But it's a really, really big part and an important part of of my job And the reason i I do this, you know, if if folks come to me
Speaker: asking for $20,000 solution, but the answer is actually something that you know could cost them $500 or something yeah a that they could do themselves. It's my responsibility to to tell them that. Because in my view, them solving that one problem that they could do themselves will help build the momentum as we've spoken before to start getting to bigger problems. And they'll come back to me, right? They'll say, hey, you know we couldn't do this with you, but actually now we're thinking this can work. so
Speaker: you know it was such a it was such an interesting experience to go through that with them because you know they they didn't have an agentic AI problem. you know They just had you know repetitive work that was distorted and wrapped in this sort of anxiety and FOMO really around around ai Yeah. i'm glad you were I'm glad you brought up AI. This story really hits home with me. And I wonder where you see AI, you know, really helping...
Speaker: non nonprofits, obviously, specifically, aside from the, oh, you know, we could build a dashboard much more quickly, or, you know, we don't have to have this little worksheet on the side, because we could have it, you know, built more quickly, whatever it is, but, but sort of more on the on the pieces that you've been talking about on reducing frictions, which could be HR pieces, or people communicating, like, where do you see,
Speaker: And maybe this differs by like the one two person nonprofit versus the 100, 200 person nonprofit. But where do you see AI having a real benefit in the nonprofit space?
Speaker: There's ah a couple of areas that I'm really interested in that I think, you know, can can add a lot of value. um The first is, you know, for those smaller companies.
Speaker: organizations and teams, AI at its very core in its base level plans can give you access to to skills, knowledge, insights that they might not have in house and didn't have access to before because they can't afford it. Right. Yeah. um we We worked with a really, really small nonprofit, ah one full time person, bunch of volunteers, and um they came to us kind of tangentially and they're like, oh, look, do you have any grant writers that you'd recommend?
Speaker: And, you know, I wanted to do my due diligence. So we we had a bit of a chat about it. And, you know, the the the the problem became quite obvious for for them and for for me.
Speaker: They couldn't afford a grant writer. You know, they're asking for it. They get the grant, right? Yeah, yeah. But they... you know What was clear is they they knew their work, they knew their their community really well, they they clearly had the expertise around it. So you know rather than looking to outsource or replace that expertise, you know we helped them to to how they could use you know just the chat GPTs, the basic plans, that kind of stuff, for how you could use AI to help them structure some of the answers to grant um grant proposals, challenge what they'd written, find holes and gaps and you know help them just get past staring at a blank page or a blank grant response. right and and And I think to me, that's that's a really good use of ah of of AI for people
Speaker: smaller nonprofits because it's not doing the work instead of you. um You know, i think that that can be problematic for a lot of organizations, but it is helping you make better use of of of what you already know.
Speaker: Yeah. um You know, which I think is really important. The second area, and and this is one i I geek out on a lot more, um is around using AI and in qualitative data.
Speaker: So, you know, nonprofits, I mean, all organizations, to be fair, but nonprofits especially, they're, you know, got loads of really, really rich information, you know. grant reports, funder reports, surveys, case studies, stuff that was, you know, captured at a workshop or a feedback session.
Speaker: And, you know, historically, we've done a pretty good job at analyzing the things that are easy to count. Right. yeah So you've got, you know, number of people served or you know, the number of workshops or the number of grants or, you know, whatever that kind of looks like. um Meanwhile, you've you've got this whole like plethora of information over here sitting sitting in, you know, the paragraphs of text that nobody really has the time to read across scale and and AI can really change that.
Speaker: um We actually did a ah big piece of work with a foundation. They wanted to look at the last couple of years of grants that they'd given out and the grant reports that were coming back, you know, the update reports. They had about 200 grants that they'd done and each of the reports, don't know, 15, 20 sort of narrative responses captured in it. And, you know, outside of someone looking at the report, nobody's looking...
Speaker: across all the reports and looking at, you know, what that's really telling them. So we used AI to help analyze those responses and really surface some of those patterns across the different portfolios, across the different geographies, so that they could kind of break all that down. And, you know, it's again, it's not using AI to make decisions about what people do. It's about how can you use AI to make that information easier for people to see and understand and comprehend and go through it. And, yeah you know, that that's helped them start to make more data-driven decisions. And
Speaker: you know, to your point around dashboards, I think that's where the world of AI and sort of data visualization gets really interesting when you put them together because you've got AI turning huge amounts of data, like unstructured information into something people can, you know, explore and and really make sense of. So yeah, I geek out about that a lot. It's, it's you know, I think it's something, it's a really sort of untapped area of of insights and ways to look at your organization and,
Speaker: you know who you're working with and why you're working with them and how well you're doing yeah um to to really drive that forward. Yeah, I totally agree. I mean, that the AI, at least for... maybe surface level or your basic analysis of qualitative data, the AI tools have opened a whole new window for people to explore that data. Now, there are lots of ways in which those tools don't get into the deeper analysis and you know the actual like deeper qual methods. But you know if you just want to pull in your eight focus groups and get like figure out what are the top line
Speaker: themes, the AI tools have been ah pretty great at that. And yeah, I can see how that can be really beneficial. And then of course you've got the whole privacy security thing, which, which, you know, iss a whole other, you know, whole other question, but yeah, I agree. i I am optimistic about what will happen with qualitative data and qualitative research because these AI tools.
Speaker: I think the trick that people are missing, and and this is something we've been deploying for a while. so you're absolutely right. If you took a bunch of information, dropped it into ah a chat GPT or a codex or a Claude or whatever, it will do a it will do an okay job of giving you some surface other stuff.
Speaker: what What we've been using AI to do is rather than say, here's a load of data, kind of just tell us what it means. We develop out quite a ah rigorous sort of code book of what we actually want to analyze. And then you give it one piece of information and ask it to analyze it just for one part of that code book.
Speaker: and give a rationale for it. So it's a really, really small task, really, really small, you know, kind of tight boundaries on that. And then you just repeat that. So, you you know, we, for the Reddit research we done, we ended up analyzing each post, I think it was about 26 times.
Speaker: So rather than all of the data look at it once, yeah understand what actually we're trying to analyze, get it to code it as it goes one bit at a time. And you know that that improved. We still had to do spot checks. We to refine the code book and stuff like that.
Speaker: But the accuracy improved dramatically for that compared to what our human coders were doing. um And again, you need automation to do that. People can't be expected to copy, paste, copy, you know all of that sort of stuff. So automation can really drive that. So I think there's definitely an opportunity to go deeper with AI once if if you get the structure right around how you're using AI. um but the temptation just to drop everything in and say, tell me what this means. yeah And it'll give you a really good answer. Like it will give you yeah ah ah a really good sounding answer, but you know, how much of it is inference, how much of it is distorted based on what, you know,
Speaker: um whatever mood it's in. Yeah, no, I mean, what you've asked it before, right? Correct, correct. Yeah, yeah, 100%. So that's why it's really important to give it a really structured task, really small structured task. And, and you know, the accuracy and the the quality assurance but behind that goes through the roof.
Speaker: Yeah, yeah. And exactly to your point, it's good for that cursory first pass. And then of course you could use that to do the exact ah ah process you just laid out, which is now we're gonna zoom in a little bit more and we're gonna do the work as humans to to go through this, but now we're gonna ask you to go into these other avenues of of interest. um David, this has been great. um Where can folks find you, connect with you? Their nonprofit needs help. Where can they seek your advice?
Speaker: Yeah, so i'm I'm very active on LinkedIn. I think we'll share the the link in the profile notes. For folks who aren't really too sure where to start, they know something they know the work's a bit harder than it should be. um We've developed out a free diagnostic tool on our website that folks can use. We'll drop the link in there. And I'd also encourage people to check out the the Reddit research. It really kind of goes into a lot of detail around giving a really interesting window into what pressure constraints and operational friction looks like in the nonprofit world through the lens of what people are posting on Facebook.
Speaker: on reddit so i think those links will be in the in the show notes for sure um like i say i geek out about this sort of stuff so very happy to have a chat with folks uh about this sort of stuff feel free to reach out anytime uh david thanks thanks john appreciate you mate take care
Speaker: Thanks for tuning in everybody. I hope you enjoyed that episode of the show. I hope you will check out more episodes. I've got more coming. And of course we've got the live podcast episode coming up at Viz in Boston in just a couple of weeks.
Speaker: And of course, I would be remiss if I didn't ask you to please rate, review the show, wherever you get it, Spotify, iTunes, YouTube, wherever you're listening or watching the show, I'd appreciate you clicking one of those little buttons. Give that thumbs up, give those five stars. It does help me get better guests and more guests on the show.
Speaker: All right. Until next time, this has been the PolicyViz podcast. Thanks so much for listening.
