Hand-knotted is a construction, not a grade. Within it the quality range is enormous, and the signals that matter are not the ones the trade quotes.
In order of importance
- 1. Wool quality
- The most consequential and least discussed
- 2. Dye quality
- Determines how the rug ages, not how it looks now
- 3. Evenness of weave
- Consistent tension, straight sides, flat lie
- 4. Finishing
- Selvedges and ends integral and well made
- 5. Knot density
- Meaningful within a type; misleading across types
- 6. Design execution
- Crisp drawing, corners that resolve
1. Wool
Compress a handful of pile firmly, hold five seconds, release. Good wool springs back almost fully and immediately. Poor wool stays compressed and has to be teased up.
Then look at the pile from a low, raking angle. Good wool has visible lustre that shifts as you move; poor wool is flat and chalky. Then feel it: good rug wool is slightly cool, dense and springy, while dead wool — pulled chemically from hides rather than shorn — feels dry and faintly crunchy.
Dead wool is legally "100% wool", which is why that phrase alone tells you nothing. The fibre science is in wool rug properties.
2. Dye
Bend the pile back along a row and compare base to tip. Natural dye is usually richer at the base and softer at the tip. Uniform colour top to bottom means a young rug or a chemically washed one.
Look for abrash — banding from batch variation. It indicates small-batch dyeing and is desirable.
Then press a damp white cloth firmly on a saturated colour. Any transfer means unstable dye and a rug that cannot be cleaned normally.
3. Evenness
Lay the rug flat. It should lie flat. Persistent waviness or curling indicates uneven warp tension, which cannot be fully corrected and is a manufacturing fault rather than wear.
Check the sides are parallel and the ends square. Sight along the length. Tribal rugs drift and that is characteristic; a city workshop rug presented as fine should not.
4. Finishing
Selvedges should be integral to the weave, not applied afterwards. Ends should be secured. These are the parts that fail first and their quality predicts how the rug will age — rug repair.
5. Knot density, in context
The most common analytical error
Comparing knot counts across traditions. A 60 KPSI Qashqai and a 400 KPSI Isfahan are not competing — the Qashqai was never attempting fineness, and judging it on density is like judging a hammer on sharpness. Density is a real signal only within a comparable type.
It is also quoted in incompatible units: raj per 7cm in Tabriz, knots per horizontal inch in Nepal, line count in China. Converting between them casually is a common way to be misled.
6. Design execution
Look at the corners of the border. In a cartoon-woven rug the pattern should resolve neatly where the borders meet. Improvised corners are characteristic of memory-woven tribal work and out of place on fine city weaving.
Check whether curves are smooth or stepped. Stepped curves in a design that wants to be curvilinear means the density is insufficient for the drawing.
Putting it together
The five-minute assessment: compress the pile, look at it raking, bend a strand to compare base and tip, press a damp cloth on it, lay it flat, check the selvedges, and look at the border corners.
That tells you more about a rug than any number a seller will quote. The full inspection sequence is in the rug buying guide, and what the answers mean for price in rug value and appraisal.
Common questions
Is knot count the best measure of rug quality?
No, and relying on it is the most common analytical mistake buyers make. It is meaningful only within a comparable type — comparing a tribal rug to a city workshop rug on density tells you nothing, because they were not attempting the same thing.
What matters more than knot count?
Wool quality, by a distance. A coarse rug in excellent highland wool will outlast and outperform a finely knotted rug in poor wool, and the difference is readable in five seconds by compressing the pile.