How AI Can Support a Business Like South Coast Stone | Practical Ideas

I've been thinking about South Coast Stone lately. Brilliant business, been supplying natural stone, paving, aggregates, that sort of thing for years. Proper quality materials, people who know their stuff, the kind of place builders and landscapers trust.

But here's the thing, running a business like that is absolutely relentless. The phone never stops. "Have you got this in stock?" "What's the difference between these two types of sandstone?" "How much do I need for a 30 square metre patio?" "Can I collect today?" "Do you deliver to Bournemouth?"

And that's before you get into the complex stuff. Matching existing stonework. Advising on the right materials for coastal installations. Working out load calculations for retaining walls. Explaining why Indian stone costs what it costs.

Every single one of those conversations requires someone who actually knows stone. You can't just put anyone on the phone. But that means your most knowledgeable people are spending half their day answering the same questions over and over instead of helping with the complicated stuff or building relationships with trade customers.

This is exactly where AI makes sense. Not replacing the expertise, you can't automate decades of knowledge about stone and masonry. But handling the repetitive queries, freeing up your actual humans to do what they're brilliant at.

The stock enquiry situation

I'd bet good money that 40% of the calls coming into a stone supplier are "do you have X in stock?"

"Have you got grey limestone in 600x900?" "Do you stock silver birch cobbles?" "Have you got any reclaimed York stone at the moment?"

Every call needs someone to stop what they're doing, check the system or physically go and look in the yard, come back, give an answer. Meanwhile three more calls are queuing up and someone's waiting at the trade counter.

An AI phone system handles this instantly.

Customer calls. AI answers. "Morning, South Coast Stone."

"Yeah, have you got any blue/black slate chippings, 20mm?"

"Let me check the stock for you... yes, we've got about eight tonnes of 20mm blue/black slate chippings at our Poole depot. How much do you need?"

"Probably a tonne bag."

"No problem, we can do that. Want to collect or shall I check delivery options?"

The AI's connected to your inventory system. It knows what's in stock, where it is, how much there is. It can handle simple stock checks, take orders for standard items, even arrange collection times.

Your staff aren't spending twenty minutes on the phone walking someone through what paving you've got in stock. They're dealing with the customer who needs advice on pointing heritage stonework or the landscaper planning a complicated job who needs proper consultation.

I've seen stone and building suppliers using this kind of setup. Cuts phone time by 50-60%. And customers get instant answers instead of waiting on hold or leaving a voicemail that doesn't get returned until the next day.

When someone asks ChatGPT where to buy stone

Right, so imagine a landscaper's working on a project in Christchurch. Client wants natural stone paving. The landscaper hasn't worked with stone much before, usually does porcelain or block paving.

They're sat in their van at lunchtime, they open ChatGPT and ask "where can I buy natural stone paving near Christchurch, and what should I know before I order it?"

If South Coast Stone's website is just product photos and a contact form, they're probably not getting recommended. But if the site's got proper content that answers real questions... completely different story.

"Here's what affects the price of natural stone paving." "Here's the difference between Indian sandstone and limestone." "Here's why you need to seal some types of stone and not others." "Here's how to calculate how much you need and why you should always order 10% extra." "Here's what installation mistakes people make with natural stone."

That kind of content means AI can confidently say "South Coast Stone in Poole would be a good shout, they've got detailed information about different stone types and they clearly know what they're talking about."

Your competitors, the national chains, the random aggregate places that happen to stock some stone, they've mostly got generic websites with no real depth. You don't need to be perfect, you just need to be more helpful than that.

And when a landscaper or builder comes to you because AI recommended you, they're already halfway convinced you know what you're doing. You're not competing on price with three other suppliers, you're the trusted expert who's going to help them get it right.

The "what stone do I actually need?" problem

This is where stone suppliers earn their money. Someone wants to pave their garden, or build a wall, or clad their house. They've seen something they like on Instagram but they don't know what it's actually called or whether it's suitable for what they want to do.

They need advice. Real, knowledgeable advice from someone who's worked with stone for years.

But here's the thing, most people don't want to make that phone call until they've got at least some idea of what they're talking about. They don't want to sound clueless. So they faff around on Google for an hour, get more confused, and maybe give up or go with whatever Wickes has in stock.

A chatbot on your website solves this beautifully.

Someone lands on the South Coast Stone site. They're browsing paving options. The chatbot pops up. "Need help choosing the right stone?"

"Yeah, I want something for my patio but I don't really know where to start."

"No worries, let's narrow it down. Is this for a domestic driveway, a patio, a path, or something else?"

"Patio."

"Lovely. What kind of look are you after, traditional or contemporary? And is it somewhere sheltered or exposed to the elements?"

"Quite modern, and it's right by the coast so it'll get battered by sea air."

"Right, you'll want something that's not going to deteriorate in coastal conditions. Limestone can suffer near the sea, but granite or a dense sandstone would be perfect. Here's a few options..."

The chatbot's not replacing the expert advice. It's doing the initial qualification and education. By the time that customer actually speaks to a human, they're informed, they know roughly what they want, and your staff can focus on the nuances and making sure they get exactly the right materials.

This also works at midnight on a Sunday when your yard's obviously closed but someone's planning their project and wants to start getting quotes lined up. The chatbot helps them, qualifies them, maybe even generates a rough quote, and then books them in to speak to someone during business hours.

The delivery question that everyone asks

"Do you deliver to [insert location]?"

Every stone supplier gets asked this about 50 times a day. And the answer's usually "yes, if you're within our delivery radius and the order's big enough" but then there's a whole conversation about minimums and costs and timing.

AI handles this trivially.

"Do you deliver to Salisbury?"

"We do, yeah. Salisbury's about an hour from our Poole depot. Minimum order for delivery there would be £200, delivery charge is £45 for standard items. Hiabs or crane jobs are priced separately. What are you looking to order?"

Instant answer. Customer knows where they stand. If they're only after two bags of gravel it's probably not happening, but if they're ordering a pallet of paving they know it's doable and what it'll cost.

And your staff aren't spending half their day explaining your delivery policy. They can focus on coordinating the complicated deliveries, the ones where access is difficult or timing's critical or you need to arrange a crane.

Making your knowledge searchable

Here's where it gets interesting for a business like South Coast Stone.

You've got decades of knowledge about stone. What works where. What lasts. What looks good. What's a nightmare to install. What's worth paying extra for. What's falsely marketed. What causes problems down the line.

Most of that knowledge is in people's heads. Maybe some of it's in old emails or quotes or conversations. But it's not written down anywhere that AI can find it.

If you get that knowledge onto your website, structured properly, two things happen.

One: When people search for stone-related questions on Google, ChatGPT, Perplexity, whatever, your content starts showing up. "Here's why porcelain isn't always better than natural stone." "Here's what to look for in quality Indian sandstone." "Here's how to avoid common mistakes with stone paving." And it's South Coast Stone answering these questions.

Two: You become the authority. Not just a supplier, the people who actually understand this stuff. So when someone's ready to buy, you're top of mind.

I've worked with suppliers who've done this. They spend an hour every couple of weeks writing up answers to questions they get asked repeatedly. AI can help draft it based on conversations or notes, then someone who knows their stuff makes sure it's accurate.

After six months you've got 25-30 genuinely useful articles. After a year you've basically documented your entire expertise in a way that both customers and AI search engines can access.

The suppliers who do this now will own AI search results for stone and masonry for the next five years. Everyone else will still be competing on price on Google Shopping.

Sample and offcut management

Every stone yard has the same problem. Someone orders stone, you cut it or they collect it, there's offcuts left over. Or you've got sample pieces sitting around. Or discontinued lines that you've got half a pack of.

Most places just let it pile up. Maybe they've got a "specials" section out the back that nobody really looks at.

AI can turn this into actual revenue.

Connect your offcuts and samples to your website. When someone's browsing for, say, limestone paving, the AI says "by the way, we've got 8 square metres of this in premium grade at 30% off because it's the end of a line, if that's enough for your project you could save yourself a few hundred quid."

Or someone's looking for stone for a small courtyard garden. The chatbot says "we've got some beautiful reclaimed York stone, about 12 square metres, it's offcuts from a big heritage project so it's not a full pack but if you can work with mixed sizes it's stunning stone at about half normal price."

You're selling stuff that would otherwise sit in your yard for years. The customer gets a deal. Everyone's happy.

The same system can manage your sample requests. Instead of someone having to physically find samples, label them, arrange postage, the AI takes the request, generates the paperwork, flags it for your warehouse team, sorts the logistics.

One builder's merchant I know set this up and started shifting about £3-4k of dead stock every month that would've otherwise just taken up space. Paid for the system in about six weeks.

The technical specification conversation

This is where human expertise is absolutely critical and AI can't replace it. But it can make it way more efficient.

Someone's specifying stone for a commercial project. They need technical data, load-bearing calculations, slip resistance ratings, frost resistance, all that stuff.

Instead of your technical person having to dig through files or manufacturer specs or their own memory every time, the AI's got it all indexed and accessible.

Customer asks about slip resistance on a particular granite. AI pulls up the exact test results, the rating, what that means in practice, where it's suitable and where it isn't.

Customer needs to know if a specific limestone is suitable for a exterior wall cladding on a coastal building. AI references your knowledge base, finds similar installations you've supplied, notes any issues or considerations, and flags that they should probably speak to your technical team about fixings and sealants given the coastal location.

You're not dumbing down the expertise. You're making it accessible faster. And your technical people can focus on solving complex problems instead of being walking filing cabinets.

Email and quote follow-up

Every stone supplier knows this pain. You send a quote. Customer says they'll get back to you. They don't. You're supposed to follow up but you're busy, you forget, or you do it once and then give up.

Most quotes just evaporate.

AI can handle follow-up sequences automatically. Smart ones, not the annoying "just checking in" nonsense.

You send a quote for a patio stone job. Three days later, AI sends an email. "Hi, just wanted to make sure you got the quote for the Indian sandstone. One thing I didn't mention, that particular stone's really popular and stock moves fast, if you're working to a deadline it's worth ordering sooner rather than later. Any questions, just shout."

No response. A week later, different angle. "Saw you were looking at the grey limestone as well as the sandstone. Here's a project we supplied last year using the limestone, might help you visualise it. Happy to arrange samples if that'd help you decide."

Still no response. Two weeks later. "No worries if you've gone another direction, but we're doing a delivery run to your area next Thursday, if you wanted to avoid delivery charges that'd be the time to order. Let me know if you're interested."

Persistent without being annoying. Helpful without being pushy. And you're not manually tracking 40 outstanding quotes trying to remember who you've followed up with and when.

I've seen this increase quote-to-order conversion by 20-30%. Turns out people often do want to order, they're just busy or indecisive, and a few gentle nudges with useful information is enough to get them over the line.

Training new staff

Stone and masonry knowledge takes years to accumulate. You can't just hire someone and expect them to know the difference between all your sandstone varieties, or what's suitable for a driveway vs a patio, or how to advise on fixing methods.

Usually it's months of shadowing experienced staff, asking questions, making mistakes, gradually building up knowledge.

AI can accelerate this massively.

New starter joins. They've got access to an AI system that's been trained on everything your experienced staff know. They can ask it questions. "What's the difference between riven and honed finish?" "Why would someone choose granite over sandstone?" "What do I need to know about selling stone for a coastal property?"

They get instant, accurate answers based on your company's actual expertise. They're learning faster. They're not constantly interrupting your senior people with basic questions. And when they do ask the senior people stuff, it's the complex edge cases that are actually worth discussing.

Three months in, your new person's operating at a level that would normally take a year. Because they've had access to the accumulated knowledge of your entire team from day one.

What this looks like in practice

Right, so let's say a business like South Coast Stone actually implemented some of this. Not everything at once, just the high-impact stuff.

Month one: AI phone system handling stock checks and basic queries. Takes maybe 30-40% of incoming calls off your team. Costs about £200-300 a month, saves probably 15-20 hours of staff time a week.

Month two: Chatbot on the website. Answers questions about stone types, helps people choose materials, qualifies leads, takes basic orders. Maybe £150 a month. Captures evening and weekend traffic that would otherwise go elsewhere.

Month three: Start documenting knowledge. One detailed article every week or two. "How to choose paving stone." "What causes stone to stain and how to prevent it." "The real difference between cheap and expensive stone." AI drafts based on your expertise, you tidy it up. Two hours a week.

Month four: Automated quote follow-up. Every quote gets a sensible, helpful sequence of follow-ups. Maybe another hour a week to set up and manage.

Month six: You've got a system that's handling hundreds of routine queries, you've got 10-15 genuinely useful articles published, AI search engines are starting to cite you, your quote conversion's up 25%, and your staff are spending their time on the stuff that actually requires human expertise.

That's not theoretical. That's businesses across Dorset, Hampshire, the South Coast doing this right now.

The bit that makes this work

You can't just buy generic AI and point it at your business.

Someone needs to actually put in your knowledge. How you operate. What stone types you stock. What questions your customers actually ask. What differentiates you from the aggregate place down the road that happens to sell some paving slabs.

It needs to sound like someone helpful who knows about stone, not a corporate robot reciting product descriptions.

But here's the thing. Your competitors, the other stone yards, the nationals, they're not doing this yet. Most of them are still operating the same way they did in 2015.

The businesses that get AI working properly in 2026 are going to absolutely dominate for the next decade. Because once you've built it, it just runs. It gets smarter the more it's used. And everyone else is still answering the phone forty times a day to check if you've got limestone in stock.

---

If you're running a stone supplies business, or any trade supplier in the South, and you're thinking "yeah this sounds good but I don't even know where to start", let's have a chat. I'll walk you through what actually makes sense for your situation. No sales pitch, just a proper conversation about whether AI's genuinely useful for what you do.

Or if you want to see what we do for businesses trying to get found by AI, have a look at our AEO services. We work with suppliers, trades, and local businesses across the South Coast, and we've spent years figuring out how to make this stuff actually work in the real world, not just in theory.