)
)
Key Takeaways
The questions changed: Clients moved from "are you using AI?" to "is it secure, is it cheaper, is a human still checking this?".
Trust beats a sales pitch: Don’t sell clients an AI story, show them real results! The good, the bad, and the unexpected.
No, it's not cheaper. But you get more for the same budget: one sprint freed up enough capacity to deliver three extra pieces of scope the client hadn't paid for.
Prototypes are easy now, running them isn't: Anyone can vibe-code a demo. Cube's value is making it survive contact with the real world, at scale, securely.
Clients are all at different speeds: Some are flying, some are just starting, and meeting each where they are is the actual job.
Lottie and Lavy spend their days between our Factory Rebuild and our clients. Going AI native didn't change what Client Success (CS) is about but it did change almost everything about the conversations the CS team had. So here's what clients are really asking, how the CS team communicates with clients, and why showing clients how we actually use AI beats any polished case study.
Meet the team in the middle
Every other blog in this series has been a team talking about how they changed the way they build or the way they work. This one's different as Client Success sit smack bang in between our Factory Rebuild and our clients. This means when the build teams changed how Cube delivers, CS were the ones who felt the client reaction land in real time.
So, introducing Lottie and Lavy. Lottie is Cube's Client Success Manager, four years in and she looks after a spread of mostly UK and rest-of-world clients. Lavy is our Client Success Director, who’s been with us for a year and who works across US and UK clients. Their jobs are to make sure client relationships are as strong as can be, and that we’re delivering the best value we possibly can.
When the internal Factory Rebuild announcement first landed, Lottie and Lavy felt what most people felt. A jolt of excitement, closely followed by "hang on, what does this mean for my clients?"
"I had that real combination of excitement, it's exciting to be on the front foot of something, but equally that trepidation. How is this actually going to work? What does it mean for my clients? How do I communicate it to them?"Lavy Shaw, Client Success Director, 3 Sided Cube
The nerves didn't last, and the reason they didn't is because CS started using the tools themselves, day to day, and trust followed from there. If you’ve been keeping up with our Factory Rebuild series, you’ll know that’s been a running theme across the agency!
"AI is running through the veins of everybody in the agency now. And the fact that we're using it ourselves has really helped onboard me, build my trust, and let me talk to clients about it a lot more myself."Lavy Shaw, Client Success Director, 3 Sided Cube
)
Every client is on a different journey
No two clients are the same, and it’s the exact same case when it comes to AI. Some are flying with these tools already, some are just dipping a toe, and some have nothing to report yet at all.
The pattern Lavy has noticed is that it splits roughly along the lines of size and age.
"The larger legacy clients are slightly more apprehensive, and that's from the standpoint of security, accessibility, all the things at the forefront of a big corporation's mind. Whereas, smaller startup clients are more agile, on the front foot, and they need their budgets to work harder,"Lavy Shaw, Client Success Director, 3 Sided Cube
There's no formal scoring system for this yet, no red-amber-green rating. Instead, the team gets the AI vibes from each client individually.
Regardless of a client’s attitude to AI, it has quickly become one of the first things CS talks about, especially with new clients.
"As part of onboarding, AI just becomes part of the conversation, and we get a good litmus test quite early of how they feel about using it. With some of the longer-standing clients, it's more of a slow burn, we've sought those conversations out to understand their appetite."Lavy Shaw, Client Success Director, 3 Sided Cube
Meeting each client where they actually are, rather than where you'd like them to be, is a big part of the job. Plus, it’s what helps you answer the harder questions well.
)
The questions changed
There's a clear before and after, and for CS it's the moment clients stopped asking whether Cube used AI and started digging deeper into the ‘how’.
"At the beginning it was a lot of 'Are you using AI?' But now it's 'Okay, you are, but is it secure? What about our sensitive data? Is it cheaper? Is it going to affect quality? Is a human looking at this before it ships?'"Lottie Dawson-Meade, Client Success Manager, 3 Sided Cube
These aren't box-ticking exercises, they’re exactly the questions you'd want a client to ask before trusting AI anywhere near their product.
Before we knew it, AI-related questions were popping up left, right, and centre across client catch ups. That’s when Lottie noticed she was giving a slightly different answer each time and we needed a way to deliver consistent answers to get everyone on the same page.
Show, don't sell
Trust is built by showing, not telling. The most powerful thing CS can do when we are talking about AI or our Factory Rebuild with clients is to pull back the curtain. To show them exactly how we use AI on our own work, what we’ve learnt, all the wins and the messy bits.
So Lottie did what any modern business professional does in 2026 and went and built a deck!
The whole thing runs through the question of ‘how might we use AI to simplify, streamline and future-proof how we build’. It starts with our mission, the same as it's always been, to change millions of lives for the better. Then it moves on to what’s working and what we’re still cracking before addressing security and data head on. All that information tends to raise a lot of questions, so to make it easier, the deck closes by explaining what we need from our clients.
“If we only showed wins, it would feel like a pitch. Showing the friction is what shows it's real."Lottie Dawson-Meade, Client Success Manager, 3 Sided Cube
Having that balance and transparency throughout the deck is intentional. It’s what turns a client update into a genuine two-way conversation and what makes them feel like real partners.
Lavy's point is that this is exactly what a client success team is here for. Walking through details with clients instead of simply telling them what's happened and what’s going to change.
"Having guardrails and talking points to genuinely take clients on the journey with us is super valuable."Lavy Shaw, Client Success Director, 3 Sided Cube
CS don't try to be technical experts either. They know their stuff, but they also know when it’s best to get other Cubes involved in the conversations. For example, when Lottie ran a session with a flood-resilience client, she looped in Guy, our Head of Product, to talk tooling.
"I wanted it to be less 'client success reassuring a client' and more a real conversation between the technical people doing it and someone who helped architect the product."Lottie Dawson-Meade, Client Success Manager, 3 Sided Cube
All of it comes back to the fact that clients can tell when you actually believe in what you're selling, especially because you use it yourself.
"When we believe in something ourselves, that shines through. We know it, we trust it, we use it day to day. That natural endorsement comes through, and clients pick up on it."Lavy Shaw, Client Success Director, 3 Sided Cube
We test everything on our own work first so we can get it to a point where we genuinely rate it, and only then do we bring it to a client. Nobody's a guinea pig.
That's the difference between "here's an AI story we're selling" and "here's the AI journey we're living."
)
Does AI make software cheaper?
The question every client asks eventually. The answer is no, not cheaper. What it does do is let us deliver more for the same budget.
"AI doesn't necessarily make things cheaper, but we can deliver more than we could have done previously. You're getting more for your money."Lavy Shaw, Client Success Director, 3 Sided Cube
That might mean extra features that weren’t originally scoped, more time to test and refine, or being able to get to a V1 of designs and then actually having room to finesse it.
This is especially exciting as most of our clients are charities and purpose-led organisations, so more delivered for the same budget isn't just efficiency, it's more impact for the cause they're there to serve.
The clearest proof of "more for your money" came from a sprint with one of our clients who's very much on the front foot with AI. The sprint had a fairly standard set of tickets like push notifications and take-action flow improvements. But because we worked AI-natively, efficiencies freed up a chunk of capacity mid-sprint.
"We went back and added three extra pieces of scope the client hadn't budgeted for. We saved time, and we gave them genuinely useful features they weren't expecting."Lottie Dawson-Meade, Client Success Manager, 3 Sided Cube
The team didn't just finish faster and down tools. Three unbudgeted pieces of scope were delivered inside the original sprint, for the original price. The client was thrilled, and got to take that win back to their own team.
Same budget, bigger impact.
If anyone can build a prototype, why hire an agency?
If anyone can spin up a prototype now, then what exactly is a client paying an agency for?
"Anyone can spin up a rough prototype now. Building one is easy. Where we come in is finessing it, making sure it doesn't fall down in six months, and knowing how to run it properly, at scale, securely."Lottie Dawson-Meade, Client Success Manager, 3 Sided Cube
Lottie's seen it play out. One of our clients loved a prototype we had built, they were inspired, and went away to try rebuilding it themselves. And they managed it, too!
But they still came back, because building the thing turned out to be the easy part. Keeping it running, secure and standing up in the real world was where they still needed our expertise.
So yes, prototypes have opened the door, but the expertise and experience are what clients actually pay for.
We've written more on the gap between a working prototype and a real product, if you've vibe coded something and wondered what happens next.
What we're still working out
Any change comes with caveats, and it’s been the same for our CS team and clients. The efficiencies of working with AI mean you can get more and faster, but it asks more of everyone.
First, keeping up with the pace. Faster delivery only works if the information comes in fast too.
"A lot of clients are super enthusiastic, but it's quick. We need a lot of information upfront, so pace is the thing that plays on my mind."Lottie Dawson-Meade, Client Success Manager, 3 Sided Cube
It's not only the speed, either. The more we deliver, the more features, the more there is for clients to review and test on their side. The extra value is great, but it does ask more of them in a development cycle than the old way did.
And it's the same story internally for us too. Claude needs a huge amount of detailed input to work from upfront, and any time our VPS (virtual private server) saves can get eaten straight back up if that groundwork isn't there. Setting the foundations first is key.
Lavy reckons the next piece of work is closing the loop with proper client wash-ups to understand how the new way of working actually felt from their side.
"We've got a really good handle on how it's working for us internally. Now it's about understanding how our clients have found it, and how we can keep improving the process from their point of view."Lavy Shaw, Client Success Director, 3 Sided Cube
)
Would they go back?
We asked Lottie and Lavy the question our Factory Rebuild series always ends on. Has all this been worth it? Would they go back to the old way of working?
Although they each have different reasons, both Lottie and Lavy agree they’re not going back anytime soon!
For Lavy, automating the admin, financial reporting, and tracking client relationships, has bought back the main thing CS runs on.
"It frees my time up for more face time with my clients, more meetings, which is ultimately what makes you tick in Client Success."Lavy Shaw, Client Success Director, 3 Sided Cube
For Lottie, the whole thing has turned into an unexpected bonding exercise, a way of getting closer to clients.
"I find the whole Factory Rebuild really exciting. We're going on this journey with the clients, it's a good bonding mechanism."Lottie Dawson-Meade, Client Success Manager, 3 Sided Cube
We’ve still got a long way to go yet, but now our clients are on the journey with us and we’re buzzing to see what comes next.
Ready for more Factory Rebuild?
Want more of the low-down on how the Factory Rebuild has gone across the teams at Cube? Get the stories straight from the Cubes. Catch up here:
You shall not pass, AI: meet the Gandalf of our Factory Rebuild
How a marketing team goes AI native (in an office full of developers)
Disclaimer: We used AI in the process of writing this blog, but a human was still at the centre! We'd be poor advocates for going AI native if we didn't use AI to help write about it. The story is ours, the quotes are our team's own words, and every line got the human pass 💚
FAQs
Does using AI make software development cheaper?
Not cheaper, but you get more for the same budget. In our experience, AI-native development frees up capacity to deliver additional scope, extra features, more refinement and testing within the budget you already had.
Is it safe for an agency to use AI on my product?
It should be, with the right guardrails. At 3 Sided Cube, a human reviews work before it ships and security checks are built into the process.
If anyone can build a prototype with AI, why hire an agency?
Because building a prototype and running a real product are very different things. Prototypes are great for demonstrating what you want but the lasting value of an agency is in finessing a product, making sure it holds up over time, and running it securely at scale.
How do I know my agency is using AI responsibly?
Ask them to show you. A trustworthy agency will happily walk you through how they use AI on their own work rather than just selling you an AI story. They should be able to explain what's working well, what they’re still improving, and where a human stays in the loop.
Published on 11 September 2026, last updated on 11 September 2026