How to sell tiles online when customers will not buy tiles online.

Tile manufacturers who build an online store and watch it underperform usually conclude that customers in this category are not ready to buy online. That conclusion is comfortable and wrong. Customers are already using digital constantly — to discover, compare, visualise and decide. What they are not doing is putting a pallet of tiles into a shopping cart and expecting the rest of a construction project to behave like an ecommerce order.

Most manufacturers think they have an ecommerce problem. What they mostly have is a mismatch between what a cart is designed to do and what a tile purchase actually involves.

The short answer is that you sell the decision, not the transaction, and route the resulting enquiry to whoever can fulfil it. The rest of this explains why that is the only version that works, and what it takes to build.

Why does the category resist the standard model?

Because tiles break most of the assumptions an ecommerce platform is built on — not one or two, a list, each item independently sufficient to cause trouble.

Two of those deserve more than a table row.

Guessing low is difficult to recover from

Ceramic and porcelain tiles are produced in lots, and the industry uses shade-variation classifications — commonly written V1 through V4 — to signal how much visual variation to expect within a product, from relatively uniform through to deliberately random.

That classification describes variation within a lot. It does not address a separate issue: material from a different lot can differ in shade or finish, because production conditions are not identical between runs. Manufacturers’ own guidance commonly acknowledges this and advises buyers to note the lot and compare current material before extending an existing installation.

So ordering short is expensive to fix, and sometimes cannot be fixed without lifting what is already down. A category where the basket has to be right first time, with a costly failure mode if it is not, is a category where a checkout flow is the wrong place to be making the decision.

A screen is not a room

How a tile reads depends on lighting, on the colours around it, on scale, on surface texture, and on the display it is viewed through. The same product looks meaningfully different in a showroom, in north-facing daylight, and on a phone applying automatic colour correction.

Online tile sellers have always known this, which is why the standard advice attached to a listing is to obtain a physical sample before committing. That advice is a quiet admission: the listing alone cannot close the sale honestly. We have set out the limits in what a visualiser can and cannot do.

What does this do to your dealers?

This is the question that decides the outcome, and it ends more tile ecommerce projects than freight and breakage combined.

Dealers, distributors, showrooms and experience centres remain central to how product is displayed, specified and sold, because the product is evaluated physically — full panels, handled samples, finishes compared under controlled lighting, laid displays rather than thumbnails. The network is not a legacy inconvenience. It is the mechanism through which the product is actually chosen, and it represents committed investment in space, stock and relationships.

A manufacturer who launches a direct channel selling the same product at the same or lower price is, from the dealer’s point of view, opening a competing outlet in their territory. Dealers respond the way any business responds when a supplier starts competing with it: less shelf space, softer recommendations, a warmer reception for the competing brand’s representative.

An online plan that does not begin with an explicit answer to “what does this do for our dealers” is a plan with a short life expectancy. It is not a logistics problem or a pricing problem. It is channel strategy.

Who is actually making the choice?

Frequently not the person paying.

For most homeowners tile is an infrequent purchase. They have no basis on which to evaluate water absorption, abrasion resistance, calibre, rectification or slip performance, and no reason to have acquired one. So they do what anyone does facing a rare, technical, expensive decision that is hard to reverse — they rely on a professional. The architect, the interior designer, the contractor, the site supervisor, the local thekedar, the mason who will lay it.

It is tempting, particularly for anyone arriving from consumer ecommerce, to read this as a distortion to be routed around. That reading is both unfair and strategically unhelpful.

The professional recommending a tile is accepting accountability for it. If the material cups, if calibre is inconsistent and joints will not line up, if shade shifts between boxes, if breakage is not replaced quickly, if the dealer cannot supply on the day the site needs it — the professional absorbs the rework, the schedule slip, the client’s displeasure, and damage to a reputation built over years.

Their preferences are shaped first by which products behave predictably on site and which suppliers respond when something goes wrong. That is not gatekeeping. It is earned, expensive knowledge, and it is precisely what the homeowner is relying on them for.

The operative point: the specifier is a customer in their own right. They have their own needs — quantity estimates they can put into a quotation, visuals they can show a client, specification data for a drawing, samples obtained quickly, availability they can rely on. A digital strategy serving those needs recruits the most influential participants in the category. One that reaches past them to the homeowner is competing with the people whose recommendation decides the sale.

Why is selling samples online harder than it looks?

Most manufacturers, having worked through the above, arrive at the same place: if bulk tile cannot be sold online sensibly, sell samples instead. Samples are light, cheap to ship, unlikely to break, and they address the screen-colour problem.

The logic is sound. But it contains a flaw that has to be designed around rather than ignored.

A sample creates a promise that fulfilment does not automatically keep. The sample represents a particular lot. The bulk order, placed weeks later, may be supplied from a different one. In a category where the whole floor has to agree with itself, a small difference is enough to cause a dispute — and the customer’s reasonable position is that they approved what they were sent.

Three honest responses, and a manufacturer should choose one deliberately rather than drifting into the first by default.

Batch-linked sampling. The sample carries its lot identity, and the order is fulfilled or reserved from the same lot. This is the version where the promise genuinely holds. It demands real inventory discipline — lot-level stock visibility across warehouse and dealer, reservation logic, and a time limit after which the guarantee lapses.

Indicative sampling, clearly labelled. The sample is presented as representative rather than exact, with the variation classification shown and lot behaviour explained in plain language. Conversion is lower. Disputes are lower still, and the buyer is being treated as an adult. For high-variation product this is often the only truthful option.

The sample as a funnel, not a transaction. Its purpose is to bring the customer into the showroom, where the material can be inspected and the order placed through the normal channel. Measured on visits and enquiries generated, not revenue. We have written separately about running a sample programme.

The reframe

Tiles are digitally decided and physically fulfilled. The online opportunity is not the transaction. It is the decision.

Almost everything before money changes hands is work digital does well: discovery, style exploration, seeing material in a room, comparing finishes, calculating quantities, producing a material list, requesting samples, sharing a design between a homeowner and their architect, checking specifications, finding the nearest dealer who stocks it.

Almost everything after the decision is work digital does badly: moving heavy fragile goods, guaranteeing lot consistency, arranging installation, handling breakage, resolving a complaint about a laid floor.

A strategy built on that division does not fight the distribution model. It feeds it.

So how do you actually start selling?

By turning the website into a source of enquiries and getting each one to the right dealer or the right salesperson, fast, without anyone sorting an inbox.

This is the part most manufacturers skip, because it is less visible than a storefront and considerably more useful. The showroom and the experience centre keep doing what they are good at. The website does the part that happens before anyone drives anywhere.

Capture the enquiry with a location attached

A pincode is the single most valuable field on the form, and the one most often left off it. Without it an enquiry is a name and a phone number that somebody has to place before they can act on it. With it, the enquiry can be assigned automatically.

Ask early — on the product page or the sample request, not at the end of a long form. Customers give a pincode readily, because they assume it is needed to tell them what is available near them. Which it is.

Route it automatically

The enquiry should reach the dealer covering that pincode, or the salesperson covering that territory, within minutes — with the product, the quantity, the project type and whatever else was captured already attached.

Three things make the difference between a routing system that works and a list nobody reads.

• Assignment by rule, not by hand. Territory, dealer tier, or who stocks the range being asked about. A person deciding where each enquiry goes is a person who will be on leave one week.
• An acceptance window with escalation. If the assigned dealer does not respond within an agreed time, it moves on. Enquiries decay quickly, and a customer who waits two days has already messaged somebody else.
• Visibility at your end. You should be able to see which enquiries were accepted, which converted, and which dealers consistently let them go cold. That data does not exist anywhere else in your business.

Follow up without depending on anyone remembering

Tile projects run for months. An enquiry that goes quiet in March is frequently a customer whose site was delayed, not one who bought elsewhere — and almost nobody follows up far enough out to catch them.

An acknowledgement immediately, a message with the catalogue or the estimate, a reminder a fortnight later, and one more after a month or two — spaced out, referencing what they actually asked about, and stopping the moment they reply. On WhatsApp rather than email, because in this trade that is what gets read.

The same channel keeps the dealer moving: the enquiry itself, a nudge if it has not been actioned, and a prompt to record what happened. See what businesses actually use the API for.

And keep doing what you are already good at

None of the above generates demand on its own. It captures and directs demand that branding, content and reputation created.

So the manufacturer’s job does not change much: brand building, design content, installation guidance, specification data, the technical answers a contractor is searching for at eleven at night. That material is what makes someone enquire in the first place, and a knowledge base aimed at end users and professionals earns enquiries a product catalogue never will.

What changes is that the enquiry now lands somewhere, gets to the right person, and does not quietly expire in an inbox.

What should the catalogue actually do?

Everything above depends on the catalogue, and most tile catalogues are organised the wrong way round.

They present collections, because that is how the price list and the brochure are structured. Customers search for a wood-look tile for a living room in a large format, or a matte finish that will not be slippery in a bathroom. Those are attributes and use cases, not collection names.

A catalogue filtered by size, finish, colour family, room and application gets found and used. One organised by collection requires the visitor to already know your range, which a new customer by definition does not.

That is a data problem before it is a website problem — the attributes have to exist as fields rather than buried in descriptions. We have written about it in why generic ERP breaks on a building materials catalogue and catalogues with thousands of SKUs.

Where does transactional ecommerce genuinely work?

Specific segments do work, and it is worth being precise about which.

• Decorative, mosaic and accent product, where value per kilogram is high enough to absorb freight and quantities are small.
• Repeat trade buyers who know the product code and need reordering fast rather than exploratory.
• Regional operators selling within their own delivery radius, where freight is controlled and a damaged delivery can be replaced next day.
• The adjacent basket — adhesives, grouts, levelling systems, trims, spacers and tools. Easier to ship, easier to reorder, better suited to a cart, and frequently ignored.
• Samples, handled with one of the three approaches above.

What is much harder to make work is a full-catalogue direct-to-consumer bulk tile storefront operating alongside, and potentially in competition with, an established dealer network.

What should you measure instead?

If the online channel’s job is to win the decision rather than capture the transaction, judging it on ecommerce revenue guarantees it looks like a failure however well it performs.

• Qualified enquiries routed to dealers and experience centres, and what proportion convert
• Response time per dealer, and the share of enquiries that go cold unaccepted
• Store visits attributable to digital origin
• Samples requested, and sample-to-order conversion
• Quantity estimates generated, which indicate a project at costing stage
• Specifications won with architects and designers
• Registered professionals actively using the tools
• Offline revenue with a traceable digital first touch

That last one settles the internal argument, and it is the one least often instrumented at the start. It is far easier to build in from the beginning than to reconstruct two years later when the channel is being defended in a review meeting.

What to build, in what order

1. The catalogue, structured properly. Everything else reads from it, and a tool built on inconsistent product data produces confident wrong answers.
2. Enquiry capture with location. Cheap, and it starts producing something you can act on immediately.
3. Routing and follow-up. Automatic assignment, an acceptance window, and a follow-up sequence that runs without anyone remembering.
4. Quantity estimation. Tiles, adhesive and grout worked out room by room, in real box counts. It answers the question customers are actually searching for, and captures the enquiry as a by-product.
5. Samples. Once you have decided which of the three approaches you can honour.
6. Visualisation. Powerful, and it needs the catalogue to be right first.

What can wait: a shopping cart. It is the part everyone starts with and the part this trade needs least.

The summary

Tile ecommerce projects rarely fail because the technology was inadequate or the customer was unready. They fail because a product with inverted freight economics, quantities that require project-level calculation, an expensive failure mode for under-ordering, colour that does not survive a screen, and a decision made by a professional intermediary was pushed through a model designed for products with none of those properties — and then measured against a target it was never able to hit.

The category does not need to be forced online. It needs digital tools aimed at the part of the process already happening on a screen: choosing, visualising, calculating, specifying and sharing — and a way of getting the resulting enquiry to whoever can actually close it. Done properly, that arms the dealer network instead of undermining it.

That is a smaller claim than “sell tiles online”. It is also the one that survives contact with the industry.

Common questions

Should we build a storefront at all?

Probably yes, but not as a bulk tile checkout. Build the catalogue, the enquiry capture and routing, the estimation and the samples first. Add a cart for the segments above where it genuinely works.

How do we launch without upsetting dealers?

Route the enquiry to them rather than taking the order yourself, hold online pricing at or above dealer retail, and respect territory. A dealer receiving a customer who has already chosen the product and knows the quantity is being given business, not having it taken.

Should we publish prices?

Some indication helps more than it costs. A customer who cannot form any view of budget assumes the worst and does not enquire. A price band, or a per-square-foot range, is often enough without undercutting your dealers.

What if a dealer does not follow up on the enquiries we send?

You need to know, and today you almost certainly do not. An acceptance window with automatic escalation handles it operationally, and the response data tells you something useful about your channel that dispatch figures never will.

Is batch-linked sampling realistic for us?

It depends entirely on whether you have lot-level stock visibility across warehouse and dealer. If you do not, that is the project — and indicative sampling is the honest option until you do.

What if we are a dealer rather than a manufacturer?

The argument applies more directly, because the enquiry comes to you rather than being routed. The harder part is that you carry several brands with different catalogue structures, which have to be merged into one usable view first.

If you are weighing this up

Most of what is described here can be built incrementally — a properly structured catalogue, enquiry capture, location-based routing to the right dealer or territory, lead management with follow-up that runs on its own, and messaging automation keeping both the customer and the dealer moving. Each piece can be added as the previous one proves itself.

We build and operate this for the ceramics and building materials sector. If you are assessing a first move online, or reviewing one that is not producing the numbers expected of it, tell us where you are and we will look at it with you — including the question of whether those were the right numbers to expect.

More on this: eCommerce development and which building materials actually sell online. For how we work with this trade, see tiles & ceramics.