A glaze test is only as useful as the note that stays with it. A promising blue that later turns grey, a satin white that suddenly runs, or a stain that fades on a different clay body can be difficult to understand if the evidence is scattered between a notebook, a glaze bucket and memory.
Clay test tiles solve this by holding several kinds of information in one small object. The tile shows the actual surface after firing, while an impressed key preserves the clay body, glaze code, application method and firing reference. Over time, a box or board of such tiles becomes a working archive: not a catalogue of ideal results, but a record of what particular materials did in a particular kiln.
The best system is not necessarily elaborate. It is one you can make again, read after several firings, and use without interrupting ordinary studio work.
Begin with a repeatable tile, not a perfect one
Choose a tile form that answers the questions you most often have about glaze. A flat square is quick to make and easy to store, but it tells little about movement. A tile with a shallow curve, ridge and textured section reveals much more: whether a glaze pools, breaks over raised marks, pinholes in recesses, or runs down a vertical face.
For most glaze work, a small upright tile is especially useful. Make it from the same clay body used for the intended pot whenever possible. A glaze can change substantially between a pale stoneware, an iron-bearing clay and porcelain. Colour response, melt, fit and surface can all differ, so a test on the wrong body may be only a rough indication.
A practical tile may include:
- a flat central area for an even glaze coat;
- a vertical or gently sloping face to show movement;
- a stamped or combed texture to reveal breaking and pooling;
- one or two small holes for hanging or threading onto a wire rack;
- a broad lower edge left clear for an impressed code.
Avoid making the tile so ornate that the glaze result is hard to read. Texture is useful when it acts as a small landscape for the glaze, not when it competes with the sample.
Photo by Colin Watts on Unsplash.
Set up an information key before making tiles
The most durable record is impressed into the clay while it is soft. Ink labels fall away from kiln shelves, pencil notes become separated from samples, and a code written with underglaze may be obscured by a dark or opaque glaze. Letter stamps, a pointed tool, or a simple set of number stamps are enough.
Keep the key short. A long line of information is difficult to stamp neatly and becomes unreadable after shrinkage. Instead, use a code that points to a written firing log or glaze book.
For example, a tile might carry:
- CB2 for the clay body;
- G14 for the glaze recipe or bucket code;
- D3 for three dipped coats;
- F6 for the firing record.
The code does not need to mean the same thing as anyone else’s. It only needs to remain consistent in your studio. Keep a master list on paper or in a digital record, with each clay body, glaze and firing given a stable name or number. If you change a glaze recipe, do not quietly reuse its old code. Give the altered version a new suffix or number so that the fired evidence remains honest.
A useful firing record notes the essentials rather than attempting to capture every moment of the kiln. Record the firing range or intended temperature, atmosphere where relevant, the general schedule if it differed from normal practice, and where the tile sat in the kiln. If cones are used, include their result in the written log. The tile need only bear the firing code; the fuller account can remain in the archive.
A simple tile record
| Tile code | Written record should include |
|---|---|
| Clay body | Name, supplier or recipe reference, colour, grog or other notable content |
| Glaze code | Recipe version, batch date, colourant or stain additions |
| Application | Dipped, poured, brushed or sprayed; number of coats or approximate thickness |
| Firing code | Kiln, firing range, atmosphere, schedule notes and tile position |
| Result | Colour, gloss, texture, movement, defects and intended use |
The important distinction is between what is permanently stamped and what is written elsewhere. Stamp what identifies the object. Write what may be too detailed to fit, or may need later interpretation.
Make a batch of tiles from one clay body
Making tiles in batches improves comparison. Roll or slab-roll clay to a consistent thickness, then cut identical strips or rectangles. A thickness that is sturdy enough to handle without becoming heavy is usually sufficient; avoid very thin tiles, which may warp easily, and very thick ones, which take longer to dry and occupy more kiln space.
Compress both faces of the slab before cutting. A rib, rolling pin or firm hand pressure helps align the clay particles and can reduce warping. If the tiles are to stand upright, bend the top section over a dowel, a batten or a folded strip of soft foam. A gradual curve is generally safer than a sharp crease.
Add texture while the clay is still soft. A coarse brush, serrated rib, comb, fabric, stamp or carved line can all work. Use the same texture arrangement on every tile. Consistency makes differences between glazes clearer. It is helpful to reserve one smooth area and one textured area, rather than texturing the entire surface.
Stamp the code while the clay is leather-soft enough to hold a clean impression. Test the lettering on a spare piece first. The marks should be deep enough to survive glazing and firing, but not so forceful that they crack the tile or distort its shape.
Dry tiles slowly and evenly. Place them on a porous board or turn them as they stiffen, especially if one side has a raised section. Bisque them with ordinary care for the clay body. A cracked or warped test tile can still sometimes offer surface information, but it is less useful as a long-term reference.
Photo by Joshua Kettle on Unsplash.
Use application zones to learn more from one tile
A single tile can hold several related tests, provided the divisions remain clear. The aim is not to crowd every possibility onto a small slab, but to show thickness and interaction in a controlled way.
One dependable arrangement is a three-band application:
- Leave the upper band as a single coat or light application.
- Apply a second coat over the middle section.
- Apply a third coat to the lower section, stopping above the foot or bare firing edge.
This shows how the glaze changes with thickness. Some glazes become richer and more opaque with extra material; others cloud, crawl, blister or begin to run. Mark the application count in the record, even if the tile itself only carries a general code.
For brush-on glaze, use the same brush and a measured number of coats. Allow each coat to lose its wet sheen before applying the next, as a heavy wet layer can disturb the one beneath it. For dipping or pouring, note the approximate immersion time and whether the tile was applied once or twice. Exact timing is not always necessary, but it matters when comparing unusually thin or thick results.
Wax resist is useful on the lower edge and back of the tile. Keep a clean bare strip where the tile will rest on a kiln shelf. If testing a deliberately fluid glaze, make the bare margin generous and consider placing the tile on a small catcher or sacrificial shard. Do not rely on a thin wax line to contain a glaze known to run.
For layered glaze tests, keep the system especially clear. Use one base glaze across the tile and apply the second glaze only to a defined upper area, stripe or edge. Record the order of application. “A over B” and “B over A” may not produce the same result.
Fire the tile as you would fire the pot
A test fired in a special corner of the kiln, on a different shelf, or in a different atmosphere may not predict a pot’s result. When possible, place test tiles among the ordinary kiln load and fire them with the same programme intended for the ware.
This does not mean every tile must occupy prime space. It means the tile should experience conditions close enough to make the comparison meaningful. If a kiln is known to vary from top to bottom, record where the tile sat. A group of identical glaze tiles placed in different positions can be valuable when learning a kiln’s character.
Let the tile cool fully before judging it. Some surfaces become easier to read in daylight, while others reveal depth under a raking side light. Handle the fired tile as well as looking at it. A glaze that appears smooth may feel dry or abrasive; a glossy surface may show fine crazing only after it has cooled.
Photo by JP Files on Unsplash.
Read the fired result with a practical eye
When examining a tile, ask what it tells you about use, not only appearance.
Look for colour variation between thick and thin areas. Notice whether the glaze settles in recesses, breaks cleanly over texture, or becomes cloudy where it pools. Check the lower edge for signs of running. Inspect the surface for crawling, pinholes, blisters, dry patches, sharp crystals or unintended roughness.
Also consider fit and function. Fine crazing, shivering at edges, or a surface that feels unpleasant may make a glaze unsuitable for certain forms even when the colour is attractive. A tile cannot fully replace testing on a finished vessel, particularly where food contact, repeated washing, thermal change or a complicated form is involved. It is, however, the first and most efficient filter.
Photographing tiles can support the physical archive, but do not treat photographs as the archive itself. Camera settings and changing light can alter colour. The fired tile remains the best reference for surface and touch.
Store the archive where it can be used
A test-tile system succeeds when it is easy to consult. Hang tiles by clay body on a wire rack, arrange them in shallow trays, or keep them in divided boxes with a written index. Store related tiles together: one glaze across several clay bodies, several thicknesses of one glaze, or a group of stains made from a common base.
A simple order is often enough: clay body first, glaze code second, firing code third. Keep rejected tests as well as successful ones. A rough brown, a blistered green or an over-fired white may save time later by showing what not to repeat.
The archive will gradually become more reliable than recollection. Each tile carries a small fired fact: this glaze, on this clay, applied in this way, under these kiln conditions. That is the quiet usefulness of the system. It does not promise certainty in every firing, but it gives each new test a place among what the studio has already learned.
Common mistakes to avoid
Making tiles without enough texture or slope
A perfectly flat tile can show colour and gloss, but it hides movement. Include at least one ridge, recess or incline so the glaze has a chance to reveal how it behaves.
Forgetting the clay body
A glaze code alone is incomplete. The same material can look markedly different on another clay, so identify the body on every tile.
Reusing a code after changing a recipe
Even a small adjustment to a stain, colourant or base recipe deserves a distinct version mark. A confusing archive is harder to use than a modest one.
Applying glaze too close to the firing edge
Leave a clear bare margin and use a catcher for fluid tests. A tile that runs onto a shelf creates unnecessary work and may damage the shelf surface.
Recording only successful tests
Unsuccessful samples are part of the map. Keep them, label them, and note the fault plainly. Their value often appears later, when a similar result returns unexpectedly.
Trying to judge every variable at once
Change one main thing per test where possible: clay body, thickness, firing or glaze addition. When several variables change together, the tile may be beautiful but difficult to understand.