AI and the software nobody will ever sell your nonprofit

0:52Scott

Yeah. Before I get into that, we need to talk about, OK, do you remember this? Do you remember your first experience with an HTML app? I

1:02Colleen

Don't remember what the project was, but I do remember being staggered because, well, you and I have worked. We've ran an agency that built websites for years and we have worked with web developers. I

1:15Scott

Mean,

1:17Colleen

We still do. Yeah, that's right. Yeah. And so like we know what it takes to hard code an HTML asset. And there's like lots of ideas for years. You and I have been talking about that every time it'd be like that's really expensive to produce because you need you need to hire somebody who's going to like custom code that idea. And so when we started to like create web like HTML assets, I was I was floored by how fast and how good we could build something.

Yeah. I

1:45Scott

Mean, I don't I'm trying to think of an analogy to explain it. It's like whatever you have in your head, you can now see like that's what an HTML as an output is. It's you have to think of HTML or code in this really truly instance is it's limitless in its expression. And so if you have the world's greatest coder and he or she can code something at the speed of light, I want this. And then it's like it's not just like it's so crazy what is possible.

And so you just explain to it. And then it comes with this very creative dynamic output. It is like having a custom app. It's like having a custom Web site. That's the mentality.

That's the kind of maybe the better modern day model is that everything that you want that you've ever thought, oh, there could be an app for that. There should be a Web site for that or whatever it is. Just go, hey, I want to see

2:47Colleen

That in like like it's a Web site and scroll around and play with it in Chrome. I think this like speaks to a principle of like human communication that I think you and I think about a lot, which is people can only take in so many so much information at once. Like when we are taking in something new, we need to hear a very small bit of information first and then we can build upon it. And one really helpful way to communicate an idea is to nest information inside boxes, hidden things. So it's like, OK, give me a little bit and then let me understand it progressively more as things unfold and give me ways to see it visually, give me ways to interact with it, because when I when you're just talking to me or I'm just reading something in a block of text at a certain point, my brain goes, I don't I don't get it yet because those neural pathways haven't like connect.

But when you can visually lay something out and make it interactive and make it something that you can like dig into, there is so much information that you get to communicate in such a short span of time and we just haven't ever had a tool like this to communicate in this way. Yeah. Yeah, as

3:54Scott

It comes to humans taking in knowledge, little bits at a time, I always thought of it as like you remember way back in the day in Nintendo, there's Legend of Zelda and the way that was they had like the map on the bottom right hand corner and it was all black. But when you started exploring the world's map that it would expose it and then you can now see

4:37Scott

Usage of this and let's talk about it is we were talking about launching a marketing campaign and I had a bunch of ideas about how this could work and so I fed it a whole bunch of just me talking, just talking about the marketing campaign. And then I gave it a couple other pieces of information that we had in terms of documentation and conversations that we had recorded and that kind of stuff and I put it all thing and then I said I need to communicate all of these ideas to the team. And then I started thinking through, man, if I am Colleen, I would want to know what I need to do. So make that very easy to do that. So make her like, oh, I can click my name and then only her stuff shows or everybody else's grays out, something like that.

This is what I'm saying to AI. And so I'm just explaining like what would be helpful in communicating those kind of things. And then it did it and I sent it to the team. What was the experience?

5:34Colleen

And it was so interesting because we've launched lots of marketing campaigns over the years and usually there's a lot of conversation. There's a whole time spent in breaking down the work into like tasks and people and all of the things. And it allowed us to have you actually on vacation when we did this most recently and you sent off the HTML and we had a like 30 minute stand up, walked through what you had sent. Nobody had any questions. Everybody was really clear on what they needed to do.

And it was we were so well equipped from that one asset. And again, it was just like you were talking to it. You were taking things strategically you'd already shaped and it was turning it into really practical output for us that we were able to get the whole thing live in a day or two. It's a great

6:18Scott

Communication tool. It's a great dreaming tool. But nobody really knows about it. Oh, you mean like pictures? There's GIFs, GIFs, however you like to say it.

There's movies. There's video. There's what are we talking about? What is HTML?

6:35Colleen

Yeah. And then it's like you have to launch a website every time you do something like this. Chrome, whatever your web browser is, already opens these files natively. You just open it locally. So it doesn't have to be like hosted online, but you also can host it online. So there's like tools like Netlify and Vercel that allow you to launch a web page in just a few.

Our friend, she just launched her business. And we gave her a few tips on using Claude Design and she was able to build a website and get it launched this weekend. And she's like, hey, check this out. This is thanks to you guys because this is, I didn't have a website. Now I have a website.

We often are

7:12Scott

Creating reports for clients. These reports are often just saying, hey, this is how things did. Trying to be as transparent but as helpful in diagnosis as possible. So it's this idea of not just telling people what happened. What we've been doing is we've always been a challenge to report just because it's always this very in a box thing.

You get a reporting piece of software and then it says, cool, when you report, this is how you have to report. And there's obviously customization modification, but it was a very low ceiling on what it could do, very low. And so because of that, it was always like, okay, all the context now has to live within our conversation. Now what's really wild is that you can just sit down with your team or yourself and just say, here's what's up. You just talk and you explain all the color between the numbers, everything that matters, just like speak to it.

Then you explain to it, I need to be able to see these things. You just kind of give it end goals. You don't give it step by step instructions. You say, I want to get to the end of this thing. Go find it. Go get it.

And you do that, but then you jam pack it with a bunch of context. And what we've been finding is, is that almost right off the rip, we're hitting a double. You know, this isn't like grand slam home run, bottom of the ninth, two outs, World Series game seven. It's none of that. You know what I mean?

Like it's none of that. This is just like, oh my gosh, dude, I just hit a double and I wasn't even playing this game before. So now I'm playing this game and I'm hitting a double. And so it's like, wow, what else can we do? We could get to grand slam home run, bottom of the ninth.

You know, we could get there for sure. But I'm saying this is what it is right now is that we are able to create these beautiful HTMLs that are interactable, that has nesting information. So it gives you the headline so you understand, but then it's like, why click more information, click more information, but it doesn't present it all to you at first because that's overwhelmed. This is the clutter boarding. We got to get that out of there.

9:28Colleen

This is, I feel like a good, like this is where I was going to go with like where, how work has changed the most over the last year. So I'm glad you're bringing up reporting. A year ago, if you had told me seriously, if you had told me a year ago that we would be having this conversation, I wouldn't have believed you. The, like the idea that we, you and I, who are not developers, who are not graphic designers, that we can be building really interactive, helpful web reports based on what we want to know.

10:38Scott

I remember growing up from like when I was like 14 to 18, I was in choir and all this kind of stuff. And I joined a band and just sang in it. And I always had these songs in my head and I'm like, just don't know how to play an instrument. And then I was like, maybe I should learn how to play guitar. So I learned how to play guitar.

And I'm like, ah, finally the songs can come out now. You know, like I have a song, but I didn't have a vehicle by which to get it out. And it feels like that now for strategists, that thing that you couldn't do before, you now can do, you know, you weren't a graphic designer before, but now you kind of have to think that you are, you know, you just need to know the process by which to acquire that skill digitally. And then once you have that skill, then you just need to follow the rules. And once you follow the rules, then you can come out with really great design that resonates because you have done all of this strategic work to set this whole thing up.

You cannot, and I stress this, you cannot just go to Claude right now and say, give me a campaign in Hindi for this company. Done. You did a tremendous amount of work to connect it, to give AI context and give it AI design. And you spent hours and hours as a strategist doing that. And then you said, make me a cool campaign for this company in Hindi. And then it said, oh, this is good.

And then you nudged it from there. Yeah, you're

12:01Colleen

Right. Yeah. All the work is in the context building and the strategy and the shaping and giving it a ton of rules that it can't do in order to get to the right thing. Here's

12:12Scott

The thing that's wild about this is that like you've built an ad creator, like you build an app to create ads, but you just did it within Claude Design. Because you just said, hey, I'm not an, I'm not a motion graphics person. So, and all of this is stuff you can just say to AI and it will say, great, I will build you this transient piece of software. Give me five minutes. And then you just play with it.

Keep going. So I want

12:40Colleen

To tie this back to giving days. If I'm sitting in an annual giving role, or if I'm sitting in like a nonprofit like marketing communications role and staring down this fall, I've got some choices I can make right now about like how I'm going to approach this. And I feel like if you're, if you're not using this year's giving day to change the way that your team is starting to approach these types of campaigns, I think you're really missing, missing out. Just quite frankly, because the whole purpose of a giving day is to energize your donor base. Like it's about that momentum and that energy around the donor base.

It's about getting loud. And the biggest obstacle most organizations have to doing that is bandwidth and marketing communications to be everywhere in with the right message, showing up in the right way. If I can do that in an afternoon with the right system, so can you. In the afternoon,

13:36Scott

You just need to know how to do it. You need to know how to set up the system because you probably have many of the tools or you could get it like truly. If you are in advancement, if you are in raising money, if you are a nonprofit who has to raise money, who is doing that via campaigns, you have everything that you need now. You don't lack anything. I mean that legitimately.

Like everything has been democratized. Everything is available to you. You can pull out a phone that has good enough video and audio to record yourself. And then for $20 a month or whatever, actually less with nonprofits with different things. And so then you have all of this availability to build all this stuff.

You just don't know how to do it. And that's actually the challenge. It's a little bit like we, you know, I can right now go to Home Depot and buy everything to build a house. I just don't know how to build a house. Like it's available to me.

And so then it's like, well, how important is it for me to have shelter? Well, yeah. And then so then I need to have somebody do that point. I'm just saying that as we get more and more into this era of intelligence, that nonprofits are going to realize that AI and AI implementation and AI infrastructure that taps into everything will become a non-negotiable in the very near future, legitimately. And this is something you have to figure out. And the sooner you figure it out, the farther you are going to be toward achieving, reaching your never-ending mission.

16:16Scott

Vibe coding for people out there, everybody knows what it is. But this is I think it's truly just talking to your computer and telling it what you want to see. That's all this is. Even I think coding is, the word coding in that feels a little bit, oh, you have to code. You don't. You don't actually code when you vibe code.

You just talk to your computer. That's literally what it is.

16:42Colleen

Vibe coding I feel like gets a little bit of hate because people are like, oh, it's not a good back end and it's not clean. And I think we got to like drill down to like what is the essential purpose of software? Ultimately, it's meant to solve a problem. It's about making work easier. It's a tool. It's meant to be more useful.

But we've learned firsthand and I think everybody should know this is like software is actually about profit. It doesn't really matter if the tool works. And so if you can if you can sell it doesn't mean that it's necessarily good software. And there's tons of really successful tools out there that are only just OK. If you if you can vibe code something and it can solve a problem for you, it's good enough.

You don't have to take it to the market like it's you can build something for yourself that solves a need. And we've done that with several things around here where we've just built the thing to replace a software subscription. And there's been no need to go back. Yeah.

17:39Scott

That's yeah. A hundred percent. And it's I mean, it's just it's really just starting to think about work differently entirely. It's just saying, do we need that thing anymore? We had a piece of software that helped us communicate with clients about modifications that needed to happen on Web site pages. You know, where they would just click on it and they would make a comment and then we would get that comment and then we would do that work.

So it was just a communication tool and it was very helpful and that's why we had it. Like just in the past several months, we have built our own vibe coded tools so we no longer have to pay for that software as a service monthly charge. And so that's these are the sort of things that you can do there. But I think more interestingly than that is that what is the software that you've never known needed existed for your nonprofit that you now can build that is the craziest piece of software maybe the world has ever seen? I mean, like just spinning wheels of like multiple different categories and then just have them all spin and they all land on like the most random thing like if you have a cat shelter that needs to collect all of.

Oh, gosh. All of the hairballs. I'm just getting weird. Do you know what I mean? Like it's like all of the different things. It's like who's going to ever make software that does that?

I see what you're

18:56Colleen

Saying. It's like, yeah, what is the hyper specific problem to your organization that nobody's ever going to make a market solution for because there's no profit in it. But you actually do need it. You and I've been doing this for ourselves. Like kind of every time we have a have a problem, we're like, wait, hold on. It's,

19:14Scott

I think the way that to think about this really truly in the modern day to how to translate this, how to translate it to like, when you say custom software, Scott, what, what do you mean? You know what I mean? Like, where in the world would I need custom software? So here's the answer is take a look at your spreadsheets. Go look at your spreadsheets and ask the questions.

Why am I tracking any of this? What is the importance of any of it? Ask the biggest questions you possibly can about the data on that spreadsheet and ask yourself what is the most ambitious thing that I would want to know on this? What would I need to bring in to connect to all of these things? Ask yourself all of these things.

And when you ask yourself all these things, you answer them. And when you answer them, you record it. And when you record it, you then take that recording plus every piece of thing that I'm talking about here and put it in a cloud and say build me custom software and it's built. That's what it is. That's how you build

20:31Colleen

Think something else you touched on is when you go to get started working with AI and you just talk to it. I think this is something that nobody knows. You and I were talking about this yesterday. Tell a story about how to be thinking about how to talk to Claude.

20:45Scott

Yeah, it's pretty strange. If you have something where it returns something where it is, let's say, an HTML and it's beautiful and it's great, but there's something on it that you look at and you go, I don't know what that is. There's something in your brain that just that's where it stops where you would say, I don't know what that is. That's a bummer that I don't know what that is. But I think in the age of AI, if your mindset can start shifting into, I can tell the person who just made that, that they need to make it differently, that you don't have to just lump it.

We feel that way because we've had decades of using software where you had to lump it. If the button's there that did the thing you wanted to do, good. If it didn't, then you have no option for that. There was no full 100%, I'm not exaggerating, customizability to your app. There just wasn't that.

But now there is, that's where the power is. And it's just, it's powered by your voice and your limitations and understanding of what's possible. And so, if you want something to be there, say you want that thing to be there. If you don't know what that means, then say, I don't know what that means. Say, hey, I'm not a tech expert, so you're going to have to give me analogies for me to grasp things.

Okay, no problem. It will do the thing that you want it to do. And then not only the thing that you want it to do, but it does remember who you are a little bit. It does a good job of, so it will remember, oh, hey, I need them to have comprehension on the forest before I understand the trees. And then they'll start explaining things.

So it needs to understand who you are. So the more you can do this, the better your experience with AI is going to be. We are at the beginning of a new era of how we need to think about work. And this new era is going to involve building custom software for your nonprofit. And when it involves building custom software for your nonprofit, you might be thinking, I have no idea how to do that.

The crazy thing is that you don't need to know how to do it. You just need to know your problems, and you need to have data. That's all. And all you do is you just give it your data, and you give it your problems, and you say, fix it. And it will start with something that will get you 80% of the way there, and then you will refine it from there. But you will start to see what is possible for your organization the moment you do a couple things.

One is just gather all the data and gather all the problems, and then just literally throw that into Claude and say, can you help? And then maybe just say, I need a live artifact to help with this. Maybe just throw that in there. And then once that happens, then you're kind of off to the races, and I don't need to tell you what to do from that moment on. Other than make it look nice.

Make it look the way you want it to look. And so you've done this in the past. I

23:38Colleen

Build a skill around my preferences for user interface, which is shorthand, everybody calls it UI. And it's really just, if you've never talked software talk, user interface is really just the skin that you see on top of a piece of software. It's like how we visualize a piece of software for human input, human visualization. And I like really clean design. I'm just like a girl who loves a gradient in an app.

I just think it looks so cool. So I like a simple gradient, and then I like a lot of white space. I want some buttons. I like a rounded edge. So just looking for other apps that look like what I like, or looking at UI elements out on the internet that I like, and build a library of options that I gave to Claude, and then build a skill around other preferences so that it's just really fast to design an artifact that I like.

Yeah,

24:37Scott

It's not hard anymore to do user interface stuff. You really don't have to know any of this stuff. You just need to have taste. You need to have an opinion. It's such an easy thing to do.

It's whatever's pleasing to you. I've seen the stuff that you've built, and every time I see it, I'm like, "That's awesome." None of it's cohesive with one another. Like if I saw one of the live artifacts you built and then look at another one, you're like, "Oh, you just totally..."

It's okay. You don't have to. But it's more of like, as long as that's how you want it, all it is is just it's a skin. It's like, "What are you