SIDE A · S2 · Ep4442 min

S2. Ep44 - Build Your Own Version of Any App With Claude Code — Without Copying It

00:00/ 41:37

Episode notes

How many of the apps you pay for every month do you actually use properly? And how many premium plans are you on just to unlock one feature?

This week, Noel explains how to use Claude Code to build your own version of the software your business relies on. Not by copying anyone's app, but by recreating the features you actually need, built around the way you work.

He shares a real app he's built that replaced tools costing thousands, how long it really took, and the subscriptions he won't be renewing (and what they cost him). He also walks through the simple process he follows before Claude Code writes a single line, which apps are a safe place to start, and the kind you should never try to rebuild.

Katie asks the questions every business owner will have: is this really something a non-technical solo owner can do, where's the line between rebuilding and copying, and is it actually worth the time? She also shares the community platform decision she wishes she could go back and make differently.

Plus Anthropic's new top model and a Claude feature that had both of us saying no thanks.

Send us Fan Mail

How to find us:

We have a free LinkedIn group (AI Automations For Business), the group is open to all.

Interested in learning WebMCP check out webmcp.academy

Check out Clyde, our multi-agent AI platform that connects to 1000+ apps and lets you build powerful automations without the complexity. Join the free Clyde Skool community to learn how to get the most out of it, share workflows, and connect with other builders putting AI to work in their businesses.

New for 2026, you can also find us on Substack, click here to subscribe and get all the latest news and updates from us.

If you would like dedicated help with your automations or would like us to build them for you then you can find our agency at makeautomations.ai

Or you can contact us via email at hello@makeautomations.ai.

Questions this episode answers

Is reverse engineering an app the same as copying it?

No. Reverse engineering means recreating the functions and features of an app based on your experience using it, not copying its source code, design, branding or name. Claude Code researches what the app does by looking at publicly available information, then builds something new from scratch based on those features.

What kind of apps should I try to rebuild with Claude Code?

Start with non-critical tools like PDF editors or viewers. Avoid apps that are crucial to your business like accounting software, because if the custom version goes wrong, it could be high risk. The best candidates are apps you currently pay for but could fix quickly if they break.

How long does it take to build an app with Claude Code?

It depends on complexity, but Noel spent about eight hours of AI processing time building a feature-rich construction app that would have cost nearly £3,000 per year to subscribe to. The brainstorming and planning phase is key to saving time.

Is it worth building my own app instead of paying for a subscription?

If you're paying around £100 per month or more for something, it's worth spending a couple of hours in a brainstorming session with Claude. Many small businesses pay for tools they barely use or only need partial features from, so building your own version for specific use cases often makes financial sense.

TranscriptRead the full transcript

Katie (00:27)

Hello and welcome back to another episode. Hi, I'm Katie, and of course I have Noel here with me today. Hi Noel, how are you doing?

Noel (00:36)

Yeah, doing great. How are you doing?

Katie (00:38)

Yeah, all good. We've just come back from a nice weekend in Exmoor with the dogs, haven't we?

Noel (00:44)

Lovely. Nice and quiet.

Katie (00:47)

Nice and quiet, and really good weather. Absolutely lovely.

Noel (00:50)

Yeah, amazing, for the time of year.

Katie (00:57)

Well, we've got a really exciting topic to talk about today. And I think it's going to, to quote you, blow people's minds, which is one of your favourite things to do with AI.

Noel (01:10)

I hope so.

Katie (01:16)

And of course, before we get into the main topic, we talk about any updates, any news in the AI and automation world that we think business owners need to know. So Noel, any updates this week?

Noel (01:34)

So there have been two updates. The big one this week is that Anthropic released Fable 5.2, which is their brand new top tier model. They also released Mythos 5.2, but nearly nobody has access to that. Only specific cybersecurity companies and the big Fortune 500 companies have access to it.

Katie (01:58)

And I mean, they're probably not listening to our podcast. If you do work for a Fortune 500 company and you are listening to our podcast, please say hello and send us an email. Even if it's nothing to do with AI or automations and you've just got an interest. It would make us very happy.

Noel (02:22)

Yeah, absolutely. But with Fable 5.2, this model has jumped to the top of all the benchmarks. It beats pretty much everything that's currently available. It's the smartest and best model out there right now. And I guess the reason it's come out now is that Anthropic have got their IPO in a couple of weeks, so I'm guessing that having the best model in the world will help them when they finally go public.

Noel (02:54)

In terms of cost, it's actually 20% cheaper than Fable 5.1 on the API, which is fantastic. And it's going to work out cheaper still, because it's more efficient in the way it uses its tokens. So it's going to spend less to get you a better result. That's what they're all doing at the moment. As each model comes out, they're making it more and more efficient, which is fantastic. So it's a decent amount cheaper than Fable 5.1, if you are using it.

Katie (03:33)

Okay, that's good.

Noel (03:37)

I'm currently using it just for complex tasks. I've been tinkering around with it this morning, having Claude Code look through some of my projects with the new model to see if it's doing a good job. You've got to try these things out.

Katie (03:48)

Of course you have.

Noel (03:53)

The other little bit that came from Anthropic last week is that you can now connect your bank accounts to Claude if you're in the US, and it can start giving you personalised financial advice.

Katie (04:08)

Oh my gosh. I don't know how I feel about that. I think "oh my gosh" is my reaction. And I guess my feeling is that I don't feel very comfortable with that.

Noel (04:21)

No, I don't either. That's going a bit too far for me when it comes to giving out personal information.

Katie (04:29)

Yeah. I also don't want it to tell me to stop shopping, or stop spending my money on certain things. And I'm like, they make me happy, leave me alone.

Noel (04:42)

Yeah. They did show an example of it where somebody was planning a trip, and they said, I'd love to go to Rome on holiday. And it was like, well, maybe you can't really afford to go to Rome. Or, you could do that, but you'd have to look at these sorts of hotels, because that's what you can afford. It's a bit boring, isn't it? You don't want to know that.

Katie (05:12)

No one wants to be told they've got to go to a budget hotel when they've been looking at five-star hotels.

Noel (05:19)

Yeah. We'll get you a business class flight, but you'll have to stay in a hostel.

Katie (05:23)

I mean, some people might be okay with that.

Noel (05:31)

Well, I know you wouldn't be. Neither would I, but anyway.

Katie (05:34)

Exactly. I was just about to say, why are you putting all of this on me? I'm the only one out of the two of us who has actually stayed in hostels, I'd like to point out. Thank you very much.

Noel (05:50)

Fair enough.

Katie (05:51)

It'd be really interesting. And I was joking about not wanting it to tell me to stop spending, but for me it's the security aspect. I still know people who won't have a banking app on their phone. But when you were saying you can link your bank account, I know of one person who would absolutely love that, to see where they could benefit from better financial advice. I mean, they're an accountant, so that's their jam. So I think for some people it's going to be amazing, and for other people it's going to be completely horrifying. So, really interesting.

Noel (06:45)

It is. The reason I wouldn't have it is that Claude and all AI models have a little statement at the very bottom of the input box saying Claude is an AI and can make mistakes. Well, I don't really want it making mistakes if it's telling me financial stuff. I guess a human financial advisor can also make a mistake, but I don't know.

Katie (07:17)

Yeah. I think if I'm looking for financial advice, right now I'm going to ask a human.

Noel (07:29)

Oh, 100%.

Katie (07:32)

But it'll be interesting to see who benefits from this new feature. It'd be interesting to hear some stories from anyone who has linked their bank account to Claude.

Noel (07:54)

Yeah, because I guess you could also create an interactive dashboard within Claude using their dashboard feature, so you could have it all tracked in there. But I don't know.

Katie (08:06)

Oh my god, I'm just so not that person. My brain does not work like that. Someone out there is going, oh my god, that's totally amazing, and to me that is literally a nightmare scenario.

Noel (08:22)

Yeah, it's all good until it goes wrong.

Katie (08:27)

Yeah, really interesting. Okay. Any more updates, Noel?

Noel (08:33)

They were the big ones, really. It's a bit of a quiet week again. Winding down for Christmas, a bit too early.

Katie (08:42)

I don't think AI ever does that.

Okay, so this week we are talking about reverse engineering apps using Claude Code. If I've just said a bunch of words that mean absolutely nothing to you... Noel, in plain English, what do we mean by reverse engineering an app with Claude Code?

Noel (09:11)

So it just means we're going to use Claude Code to go off and create apps that we use on a daily basis and already pay for. We're going to build them based on the features of whatever you pay for currently. That's it in plain, or plainish, English.

Katie (09:32)

Okay. So how does this differ from copying it? I don't quite want to say hacking. How is it different from copying it?

Noel (09:48)

It is quite different from copying, because obviously you won't have access to the source code for that particular product or app that you use in your business. Claude Code can't go off and completely rip off somebody else's source code and create a copy of it. That's not how it works.

Noel (10:10)

What we're doing is saying, look, I want to build an app. You can name those apps in your prompts and say, this is what I'm currently using, I really like these features. And then it can do some research into what that app does and what those features are, and then create an app for you. So it's a step away from completely copying someone else's work.

Katie (10:36)

Okay. So what got you thinking about doing this? Was there something you saw, or a particular subscription that made you think, I could just build this?

Noel (10:51)

So I've been building with Claude Code for, blimey, it's been a while now. I can't remember exactly when it came out, but ever since it came out, let's put it that way. And it's getting better and better to use with the new models. The apps I'm starting to build today are light years ahead of what I thought could be possible even a year ago, or even six months ago.

Noel (11:22)

But I started to see people completely redoing apps from Adobe, for example. Obviously those apps cost a lot of money on monthly subscriptions, and they said, we've gone off and recreated all of those, and we're now open sourcing them, just download them for free. And I was like, hmm, okay. That's what got me thinking.

Noel (11:45)

And in my electrical world, the apps that I have to pay for are blooming expensive. Do I need every feature? No, I don't. So that's where I'm coming from. Working out the expenses, the really big apps I pay for, and then how I could recreate those using Claude. And thankfully it can be done.

Katie (12:07)

Okay. And is this actually something that a solo business owner, or someone with a small team, can do?

Noel (12:20)

Yeah, it definitely is. Like we say with anything with AI, all you've got to do is ask it, and it'll start getting you going in the right direction. You don't need to be some sort of software whizz to create your own app, especially if you're using it internally. If you're just going to have it on your computer or laptop, it doesn't matter if it crashes, it's not the end of the world. And if it does have any issues or errors, you can just ask Claude Code to go off and fix them for you. So anyone can do it.

Katie (12:56)

I was going to say, how much do you need to know about coding to be able to achieve that?

Noel (13:04)

I would say not a great deal. In my workflow, I start off with a normal Claude chat and use that as a brainstorming session. I talk to it and say, look, this is my idea, this is the direction I want to go in, and we have a back and forth about it.

Noel (13:25)

Then at the very end, I ask it to create a document that I can pass over to Claude Code as the developer. It condenses everything we've talked about, and it'll even include code snippets and all that sort of stuff that would be helpful for Claude Code, if you ask it to. Then you fire up the Claude Code session, copy that document in, and say, build it. And off you go, and hopefully it makes a good start.

Katie (13:55)

So what sorts of apps are best for building your own version of? And are there any that you should never try to rebuild or replace?

Noel (14:09)

I guess anything that's in any way critical to your business, let's not do those. Let's say you've got an accounting app for your business. I wouldn't go off and reverse engineer all of the tax calculations that go into those platforms. Although yes, it could be done, that's high risk if you get it wrong. So I would leave those well alone.

Noel (14:35)

It's more the apps that we use day to day where, if it went wrong, it's not the end of the world. Or if it did go wrong, you could go off and fix it in Claude Code. So it can be for almost anything, just not something that's critical to the business, I would say.

Katie (15:01)

Okay, that makes sense. I think a lot of people listening will be wondering, how do you work out whether it's worth reverse engineering an app? Once you factor in time, hosting and maintenance, is it actually going to save you money?

Katie (15:30)

But I guess, is the answer always saving money? Because sometimes it's about getting something that actually fits your business better than the apps that are available. I've definitely been in business and gone, well, if I could merge this app, this app and this app together, because I want this bit of it but I don't want the rest of it, and I want a tiny bit of this one but I don't need the rest of it, and I want this, this and this from the other one.

Noel (16:04)

Yeah, that's actually a great use case, where you look at the different things you're currently paying for and then bring them all together. That's where it becomes super powerful when you're creating those sorts of things for your own business, because it's going to help you do stuff quicker and in the way that you like to work. With a lot of these apps you might think, right, that's 10 clicks to get to the bit I need. Whereas in your version, it could be one click, or whatever you require.

Noel (16:45)

When it comes to cost, it is kind of tricky, because it will take a bit of time to go off and build it. But if you're paying anywhere in the region of a hundred dollars a month for something, it is worth spending a couple of hours having that brainstorming session with Claude. And once you get to the code section, you can just set it off to do its thing, go and have a cup of tea, have lunch, and come back and see what it's done.

Noel (17:15)

It might need permission to start the next piece of work or whatever. That's all you're doing. You click the button and go, yeah, that's fine, then go on to the next bit, or run the app and have a look. And then you can redirect it from that point. So there is a bit of time going in upfront.

Noel (17:31)

But there's also the time you'll save from trying to learn all the different platforms. Would it be easier for you to learn your own platform that you've built? Because you could just ask Claude, how do I actually get to this piece? And it would read the code and go, these are the exact steps. Right, great, thanks, I now know how to use it, which is really awesome.

Katie (17:55)

Yeah. Because for me it was about where to host my community. At the time, Skool didn't exist, or if it did, it wasn't very well known. I was looking at things like Mighty Networks, Circle and Heartbeat. Each of them had something I liked, but what I thought was quite a basic standard feature was, no, no, no, that's going to be on the premium subscription. So then you're paying a heck of a lot of money just to have one particular feature that you wanted, and it's not always worth the premium price.

Noel (18:54)

Yeah, you're unlocking 20 features just to use one, aren't you? It's not ideal.

Katie (18:58)

I know. And I really wish this had existed four or five years ago, because I absolutely would have gone, do you know what, I'm going to create my own app where I can host my community. These are all the features I'm going to have, because this suits the type of community I have, or the type of businesses that are going to be using it.

Katie (19:30)

So I think for people who want something for a particular industry or a particular reason, this is going to be such a game changer. And that's why at the beginning I was like, to quote you, Noel, this is going to blow your mind.

Noel (19:50)

Absolutely. It is crazy, isn't it? Four years ago this would have been complete nonsense. You would never have dreamt of being able to do it. It would have cost you a fortune, wouldn't it?

Katie (19:58)

Yeah. Because I have friends who have had their own apps built for their businesses, and it has cost them £10,000 upwards. And that's just to have it built. If they want any changes or maintenance, it's even more money. So that wasn't even an option for me.

Katie (20:25)

And my friends who have had their apps built, it wasn't overnight. It took months and months. There was a lot of going back and forth, and that is just something I was totally not interested in.

Katie (20:47)

And I spent a lot of time researching all these different apps. So it's like, I'm spending so much time researching, when actually I could go, do you know what, I want this app, these are all the features I want it to have, build it.

Noel (21:04)

Yeah, that's exactly what you can do right now. It's just as easy as that, really. It's just identifying what the features are and then asking Claude to replicate them.

Noel (21:16)

And I wouldn't say they're complete copies, obviously. That's not what it's going to do for you. You can then say, actually, with that feature, if you tweaked it a little and it did this, it would be even better. And that's really powerful.

Katie (21:35)

Yeah, because I think that's what we need to say, isn't it? We're recreating the function of an app. We're not copying its design. We're not copying the branding. We're definitely not copying the name or the content.

Noel (21:50)

No.

Katie (21:52)

If you don't want to get sued, or go to court, do not just copy an app like for like. That's not what we're saying.

Noel (22:07)

Absolutely. Don't get in trouble.

Katie (22:09)

No, we don't want you to get into trouble. So, talking about not getting into trouble, should people check the terms of service of the original app they're trying to recreate before they start?

Noel (22:27)

You could, yes. But what we're saying in this episode is that we're recreating the functions and features. We're not sending screenshots of apps and going, hey, completely copy this, because that would be flagged somewhere in those terms of service.

Noel (22:48)

So I'd definitely triple check it. But as long as you're recreating features and not making a carbon copy, we're all going to be good. Because, as an example, you could cancel your Canva subscription today and you would still know what features you'd want to put in, if you were going to create your own version.

Katie (23:12)

So we won't have the source code for the app we're trying to recreate. What are we actually using when we're trying to rebuild an app?

Noel (23:27)

I guess it's sharing your experiences and the features you'd like within your particular app. That's what we're aiming for. But obviously, if it's a well-known app, all the AIs are going to know about it anyway. They can go and have a look at the website themselves. They might even be able to find screenshots of what it looks like today in different blog posts about it.

Noel (23:55)

So it can do that research and further understand what you want to build. If you say, I just want a PDF editor which is similar to Adobe, it'll go, okay, I'll go and have a look at Adobe, what have they got, what features are in there? You probably don't even have to list all the features, it can go off and find them itself. And then, right, now we want a new PDF tool, how do we start building this together and create it from scratch? That's what it's doing.

Katie (24:32)

So I'm guessing it's like anything with AI. When you're talking to your LLM, the more information and the more detail you give it, the better the outcome.

Noel (24:42)

Yes, definitely. That's why I always do that brainstorming session at the start. I'm getting all of that data and information, and at the end I'm creating that design document that I can pass over to Claude Code. It saves me so much time. It means I don't have to re-explain it all to Claude Code. I ain't got time for that.

Katie (25:07)

Okay, that makes sense. So have you got an example of something you've reverse engineered this way? Give us some details of what that looked like, maybe how long it took, and maybe compare what a subscription would have cost you with what it's going to cost you now that you've built the app yourself.

Noel (25:44)

Yeah. So recently I've created an app with two other guys called Take Off, and that's in the construction sector.

Katie (25:54)

Plug.

Noel (25:58)

Very true. It's in the construction sector, and what it relies on is plans for buildings, so we can put those into our app and it can go off and work out the cost and price, because that's what our app does.

Noel (26:15)

There are many, many tools out there. I was looking around them, and Autodesk Revit pops up, and there are similar tools in AutoCAD and all those sorts of things. But at the minute, I think they cost around £2,000, well, nearly £3,000, per year for that particular tool. And it's like, I don't really want to pay that, and then have to learn it.

Katie (26:44)

CAD tools are expensive.

Noel (26:47)

Megabucks. It's silly, isn't it? And then I'll see another app that's a bit cheaper, but it's still £89 per month. And I thought, I just don't have the time or inclination to go off and learn that system. Is it actually going to give me what I want at the end of it? I don't know. I'd have to pay and find out.

Noel (27:10)

So I used Claude Code to get all those features together and create the app. And what I've ended up with today is actually something we could probably add to our product sheet, to be honest. It's really feature rich. I can just give a floor plan to the AI and go, right, this is the ground floor plan, go off and create it. And it creates all the plans, the electrical plans, the building, how the bricks are all formed, how the roof works. It creates all of that detail.

Noel (27:45)

And that app, I would say, didn't take me long in my personal time to create. But in terms of AI time, maybe eight hours. There were a few long sessions in there while it was going off and building, because there were a lot of different features that needed adding in.

Katie (28:13)

And obviously to make sure it worked as well, and that it was actually accurate.

Noel (28:16)

Exactly. And what I've got today just baffles me, really. I was working on it earlier, and it had got to a point where I was just using the AI to create the buildings. So I said to Claude this morning, you know what, I'd like to be able to have a go and do it manually. Could you create a help guide section with animations to show me how to do the basics? So if I need to edit anything or create something, I can watch those and read the guide and know how to use it.

Noel (28:50)

And it did that in about 10 minutes, if that. Now I've got an 18 step guide on how to use this tool that I've built. It's crazy. It's completely replaced any notion of me paying hundreds of pounds a month to another provider. And the information is really good as well. I know I'm definitely going to get the output I need for the other app, because I can make sure Claude Code adds that feature in.

Noel (29:26)

So that's one example I've done recently, but I've got a few others in the works for the electrical side of things. Those ones are megabucks, so it'd be nice to recreate those.

Katie (29:42)

If someone is listening to this and thinking, do you know what, I'm going to give this a try this week, where would you suggest they start? What's a small or low risk app they could start with, just to dip their toes in the water?

Noel (30:01)

I guess we've already mentioned it, but something like a PDF editor or PDF viewer. Something that's not going to be critical to you or your business, but is enough to get your feet wet and see how it works, and start working with Claude and understanding how it works.

Noel (30:25)

Because it's one thing asking it to do things. I've been using it for so long now that I can almost second guess what it's going to come back with. So I'm learning how Claude works as well as Claude learning how I work. It's interesting. But I'd definitely give something like that a go.

Katie (30:41)

Yeah, just to give it a try. You don't necessarily need to want to use it. It's just that you know what you're building, and it's something relatively risk free, so if you don't get it completely right, that's fine. And then once you've got a bit more confidence and you know what you're doing, you can try something a bit bigger, or something you'd actually like to use in your business.

Noel (31:16)

Definitely. My Adobe subscription's finished now, so I can't edit PDFs anymore. I'm going to have to get Claude on the go so I can start editing PDFs again.

Katie (31:27)

So where do you think all of this is heading? Do you think more small businesses, or businesses in general, will build more of their own tools in the future?

Noel (31:40)

I think companies are going to start building their own stuff more in the future. Obviously people are going to keep an eye on risk. The balance is going to be, is it going to cost us more to maintain our own app than to pay somebody else? But yes, definitely.

Noel (31:58)

I could definitely see a lot of small businesses having access to what I'd call an enterprise-style product that they've just coded in a couple of days. If you could imagine having all of that in your business from day one, that would be fantastic. Because there are so many tools I wish I could have afforded at the time. Whereas now I can just go, you know what, I'll build my own, it won't take me long. So I could definitely see it happening.

Katie (32:30)

Like I said, I really wish this had been around four or five years ago. I feel like it would have been a complete game changer for me.

Noel (32:39)

100%. That's fantastic.

Katie (32:44)

Okay, I've got one more question for you, and it's up to you whether you answer it or not.

Noel (32:51)

Okay.

Katie (32:52)

What subscription have you cancelled because you've built your own version?

Noel (33:01)

What have I cancelled? I haven't actively cancelled anything. But what I won't renew are, and I hope they're not listening, my electrical design tools. I won't renew those. I'll go off and create my own.

Katie (33:20)

And how much per year do you think you're saving on that? So I want you to think about what the subscriptions for those tools cost, and then the cost of building, hosting and maintaining it yourself.

Noel (33:43)

So for the electrical stuff, there are two apps that I really like, and one of them has gone mega expensive, but I like the features in both. Combined, for a year, you're looking at around nine thousand pounds.

Katie (34:03)

Okay, yeah, that's quite a lot.

Noel (34:06)

It is a lot. And how much would it cost me to recreate them? I guess a couple of hundred pounds, maybe, in terms of usage. It's not going to be a huge amount. And I'd be able to create features that these apps aren't going to create. Because I know one of them does very little on the development front. There's so much more they could do, they just don't do it. So I'm thinking, how would I envisage that working? How could we make it even better? That's what I'd like to work on, and I've already started working on it in the background. That would save me an absolute ton of money.

Noel (35:01)

And at the end of the day, it's not actually going to cost me anything to host either, because it's all stuff I'd have installed on my laptop or PC. So I don't have to worry about hosting it in the cloud and using it there. That doesn't bother me at all for these particular apps, and for most of the apps you're probably thinking of building yourself, it'll be the same. So I won't really worry about that just yet.

Katie (35:29)

So apparently the average SaaS prices, that's software as a service, are rising by 8% to 25% a year. So the tool you signed up to at £20 a month last year or the year before probably isn't £20 anymore. Everything is going up.

Katie (36:05)

And to weigh this all up, it's estimated that 94% of businesses say they have at least one tool that they rarely use but still pay for.

Noel (36:20)

We can all relate to that.

Katie (36:24)

And here in the UK, Fasthosts have estimated that SMEs could be wasting up to £10,000 a year on unused software, with nearly 40% of workplace software going unused. And there are also the hidden time costs. The average business spends 22 hours setting up a significant new software tool.

Katie (36:55)

When I've done contracting for businesses where I've had to go in and use their tools, and then they go, actually, we're going to update this particular tool, they've then got to train all of their staff on it.

Katie (37:13)

I can remember one particular new tool they rolled out, because people didn't particularly like the previous tool and weren't updating it. So they were like, we'll buy this new fancy tool and get a year's subscription. And they found that the percentage of people using the new tool went down even further than with the previous one. So about six months into using the new tool, they reverted back to the previous one.

Noel (37:55)

That's crazy, isn't it?

Katie (37:56)

So not only have they lost a lot of money, they've also lost a lot of time, because of the training. So it's really, really interesting.

Noel (38:11)

Yeah, the training is a real big one. When I was working at a company and we brought in a new electrical design tool, everybody had to go off and do the course, and that took five days.

Katie (38:24)

Whoa.

Noel (38:26)

Five highly skilled engineers going off to sit in a classroom for five full days. That in itself was a big expense to the business.

Noel (38:40)

But I guess the way you get around that is, if Claude has built the app for you, like I did earlier with the guides in my app, it can show you how to use it. It can create fully animated videos showing people exactly how to use a tool, so they can sit and go through it in their own time. Not death by PowerPoint, with somebody pointing at a screen going, now click here, and then click here. It gets a bit hard work, doesn't it?

Noel (39:09)

But it also helps, because as the app updates or you add new features, it doesn't take much to say to Claude, great, we've added this, now add it into the training material. And it'll just go off and do that for you, and it'll be sat there ready and waiting. So it's a huge saving.

Noel (39:32)

And like you were saying, if you create an app, is it just going to die? Is anybody going to use it? Well, if you've created it yourself and you don't use it, that's fine. It just sits there on your laptop. It's never going to go away, but you're not paying for it to be there. You're not forgetting about a subscription you haven't cancelled. So there's definitely a saving to be had.

Katie (40:04)

Yeah. And I would encourage people who are listening to do a quick audit. Go through what you're actually paying for in apps or software. Are you actually using them, or only using them a couple of times a year? Maybe you're not using all the features. And see if it would be worth building your own version or your own app instead.

Katie (40:41)

It'd be really interesting to hear from people as well if you're going to be cancelling subscriptions and trying to build your own app. If that's you, do let us know. Send us an email, hello@makeautomations.ai, or come and let us know in our free LinkedIn group, which is available to anyone to join. It's called AI Automations for Business, the same name as the podcast, so hopefully you won't forget that one.

But we really hope you've enjoyed this week's podcast episode. If you've got any questions, send us an email, and we look forward to catching you next time for another episode.

Want this for your business? Let’s talk it through.Book a session · £150