Request free Shaw sample boards for your next flooring selection — soft oak LVP, carpet tile, or hardwood.
Shaw flooring article feature image

Shaw Journal

Porcelain Tile Specification Guide: Why Your Spec Is Probably Incomplete

By Emilia Novak

Most sourcing conversations start with the tile. Someone sends me a photo of a mosaic backsplash or a porcelain slab and asks, “Can you match this?” Or they skip the tile entirely and ask, “What thinset should we use?” Both questions miss what I check first: the specification.

I’m a quality/compliance manager at Shaw Flooring. In a typical year, I review about 200 incoming material orders—or rather, closer to 220 when you count pre-production samples. In Q1 2025, I rejected 8% of first deliveries. Maybe 7.5%, I’d have to pull the report. The point stands: the spec was wrong, not the product. That number isn’t about being picky. It’s about the difference between a product and a specification.

The Surface Problem: You’re Choosing Tile Before You’ve Chosen the System

The tile is the last thing I worry about. That sounds strange coming from a flooring guy, but it’s true. If the tile is defective, you can see it. If the specification is defective, you don’t see it until after installation—when it’s expensive to fix.

Porcelain tile isn’t one product. It’s a family of products defined by water absorption, dimensional tolerance, breaking strength, and surface properties. ISO 13006 classifies fired ceramic tiles into groups; porcelain is often Group BIa, with water absorption not more than 0.5% by weight. That low absorption is why porcelain works in demanding applications. It’s also why ordinary adhesive won’t always bond the way people expect. A dense tile doesn’t pull moisture out of the mortar the way a more porous tile does.

In Q1 2025, our internal audit found that 22% of tile-related field failures were tied to adhesive selection, not tile defects. The tile was fine. The mortar didn’t match the tile class. And once the tile is down, the mortar is the one thing you can’t see. (Source: Shaw internal quality audit, Q1 2025.)

That’s why “What thinset should we use?” is the wrong first question. The right first question is “What tile class am I specifying, and what substrate is it going onto?” Until you answer that, any adhesive recommendation is a guess.

The Real Problem: Nobody Owns the Specification

Here’s the part that changed how I think about my job. Everything I’d read in product catalogs suggested that choosing a good tile was the key to a good floor. In practice, the tile was rarely the weak link. The specification process was.

A commercial project spec passes through many hands. Architect specifies a tile from a cut sheet. Distributor substitutes a “comparable” product based on color and price. Contractor orders the adhesive their supplier stocks. Installer uses the trowel size they’ve always used. No one verifies that all four decisions are compatible.

The conventional wisdom is that a good tile prevents callbacks. My experience with 200+ orders suggests otherwise. A mediocre tile with a correct spec often outperforms a premium tile with a broken spec.

I didn’t fully understand this until 2023, when we received a batch of 5,000 square feet of tile-ready underlayment panels—5,200 with the overage, actually. The panel carried a product name that traced back to a distributor’s sales sheet, not a manufacturer’s data sheet. (Think of a search phrase like “Shaw Lane VIP board”: it sounds specific, but it doesn’t tell you what the product is. The data sheet does.) The vendor claimed it was “within industry standard.” Our standard is tighter, and our performance test showed edge swell of 2.3 mm under humidity against our 1.5 mm tolerance. We rejected the batch.

I still kick myself for not demanding the technical file before the purchase order. If I’d required it earlier, we’d have caught the problem before it became a $22,000 redo and a four-week delay. Now, every contract includes an edge-seal and moisture-resistance specification.

Mosaic Tile OEM: Where Specification Gaps Become Visible

Mosaic tile OEM is a different animal. When you source a custom mosaic, you aren’t buying tile. You’re ordering a system: tile, mesh backing, sheet pattern, edge type, adhesive compatibility, and lot tolerance. A missing element can ruin an entire production run.

In 2022, I implemented a pre-production sample protocol for OEM mosaic orders. It came after a supplier changed the mesh backing on a large order. We’d approved a prototype with white mesh; the production batch used a darker mesh to save a few cents per sheet. Installed with light-colored grout, the mesh shadowed through the joints. The tile itself was fine. The backing wasn’t. The product matched the catalog, but it didn’t match the spec.

That defect ruined 8,000 mosaic sheets—not in storage, but during installation. We caught it because the installer noticed the shadowing on the first wall. The supplier replaced the sheets, but the damage to the schedule was already done.

Now, if someone asks me about mosaic tile OEM sourcing, I tell them to write down four things before picking a design: actual tile dimensions and tolerance, tile thickness and edge profile, sheet size and joint width, and mesh color and composition. If the supplier hesitates on any of those, that’s a red flag.

The Cost of a Broken Specification

Let me put the cost in plain numbers. From my own records, not an industry average:

  • A $22,000 redo and a four-week launch delay from an underlayment panel whose technical file was missing.
  • 8,000 mosaic sheets replaced because an unapproved mesh color changed the appearance after installation.
  • 7% of first deliveries rejected in Q1 2025 due to specification mismatches. The products weren’t defective. The specs were.
  • Customer satisfaction scores increased by 34% after we upgraded our specification review process in 2023.

Not all of these show up as line items. The worst cost is trust: the contractor loses faith in the product, the distributor loses credibility, and we lose the next order. That’s hard to quantify, but it’s real.

The Porcelain Tile Specification Guide I Actually Use

Here’s the shorthand version. It doesn’t make projects more expensive. It makes them more predictable.

  1. Start with project conditions: substrate type, moisture, traffic, exposure, installation method. Not the color.
  2. Define the porcelain tile class you need: water absorption, dimensional tolerance, breaking strength, slip resistance. If you need an exterior floor, say so early.
  3. Source tile adhesive after the tile class, not before. For tile adhesive sourcing, give the supplier the tile’s absorption, the substrate, and the environmental conditions. Ask for the mortar’s open time, cure time, and recommended trowel notch. If they can’t answer, find another source.
  4. For mosaic tile OEM, require a pre-production sample plus a technical data sheet. No data sheet, no order.
  5. Treat product names as clues, not specs. If someone refers to “Shaw Lane VIP board” and can’t produce a data sheet, don’t assume it’s tile-ready. Test it for the actual application.

Those details come from product standards I rely on: ISO 13006 for tile classification, ISO 13007 for adhesive classification, and ANSI A108 for installation guidelines. Standards change, so verify the current version before you write them into a contract.

The Honest Limitation

Here’s the part I don’t say enough in product discussions: no single tile, adhesive, or underlayment is right for every project. I work for a flooring manufacturer, and even I don’t recommend one “best” system. If a supplier tells you their product handles every substrate and every condition, they’re offering a shortcut, not a specification. What I do recommend is that you get the spec right before you get attached to the design.

The next time a tile project goes sideways, don’t start by blaming the tile. Ask who owned the specification, whether the adhesive was matched to the tile class, and whether every product name was verified against a data sheet. That’s where the real problem—and the real fix—lives.