CRI for Retail Display Lighting: More Than Colour Temperature

1. Define the retail colour task

List what the lighting must reveal: skin tones at a beauty counter, reds in food or fashion, timber finishes, neutral greys, metallic textures or a printed brand colour. Different displays can weight these observations differently, so a single generic threshold cannot replace a project brief.

Separate colour appearance from brightness, uniformity, glare and energy use. A source can meet a colour-rendering requirement yet fail because the display is uneven or visually uncomfortable. Conversely, a bright display does not prove accurate colour rendering.

Include the face, diffuser, graphics, shelf finishes and ambient store lighting. These materials and surrounding sources can alter the installed appearance even when the light-source data is sound.

2. Keep CCT and colour rendering separate

The CIE defines correlated colour temperature as the temperature of the Planckian radiator whose perceived colour most closely resembles the stimulus under specified conditions. In practical terms, it helps describe whether a white source appears warmer or cooler; it does not say how faithfully coloured objects will appear.

The CIE colour rendering index compares the colour rendering of a test source with a reference illuminant. The general index Ra is the average of the first eight special colour rendering indices. Sources with the same CCT and similar Ra can still differ in individual samples, spectral distribution and visual outcome.

Do not use CCT as a proxy for CRI or CRI as a proxy for CCT. Specify both where both matter.

3. Build an exact-source colour evidence table

Collect data for the complete module, strip, luminaire or lamp variant, not only the LED package family. Record the document revision and tolerances. If a field is not published for that exact variant, write “not documented” and request evidence.

4. Know what Ra does not settle

Because Ra averages eight results, it can hide a weak result for a colour important to the display. It also does not describe colour preference, saturation, gamut shape, chromaticity shift between products, flicker, optical distribution or how a coloured face modifies the source.

The CIE position statement on colour quality metrics explains that no single metric can describe every aspect of colour quality and that appropriate measures should be chosen for the application. IES TM-30 provides fidelity and gamut information, but its presence does not remove the need to inspect the actual retail task.

Avoid claims such as “high CRI makes every colour accurate” or “R9 alone proves retail quality”. Treat every metric as evidence for a defined question.

5. Do not infer missing data from Glowfy listings

Current Glowfy module pages identify options such as colour temperature and selected output data, but the reviewed listings do not publish a complete exact-SKU set of Ra, R9, TM-30 and spectral data under one common method. This article therefore makes no unsupported CRI claim for K2, K8 or another Glowfy product.

A Philips 2835, OSRAM 2835 or another chip-family reference does not by itself establish the colour rendering of the finished module. Phosphor selection, binning, drive, optics and assembly variant matter. Request the full product evidence before a colour-critical commitment.

If documentation cannot be obtained, a controlled physical sample can support a project decision, but it should be recorded as a visual acceptance of that sample and batch—not converted into an invented CRI value.

6. Worked example: compare two 3000K display options

A shopfitter is choosing between Source A and Source B for a timber-and-red-textile display. Both are labelled 3000K. Source A has a documented Ra value and supporting data for the exact assembly; Source B states only CCT. The correct comparison table marks Source B’s CRI, special indices and TM-30 fields as unknown rather than assigning typical values.

Build equal display samples with the same shelf, background, diffuser, source position and target illuminance on the merchandise plane. Place the same timber, saturated red textile, neutral grey, skin-tone reference and approved brand-colour sample in each. Control other store lighting and adaptation time, then assess in a randomised or masked order where practical.

Record meter readings, observer roles, visual differences, photographs under fixed manual camera settings as supporting evidence, and the pass criterion for each priority material. The team may accept A, request documents or a repeat sample for B, or reject both. CCT equality alone is not a pass.

7. Use a retail colour-assessment worksheet

At the top of the worksheet, state the display purpose, priority colours and acceptable white appearance. Identify the exact source, variant, batch if relevant, power/control state, optics, diffuser, mounting and document revisions.

For the sample, record target and measured illuminance, locations, surrounding light, background, observer positions and all merchandise references. Keep the electrical and photometric conditions equal between candidates.

Conclude with separate decisions for documentation, visual colour acceptance, brightness/uniformity and installation compatibility. Glowfy can help arrange a sample or review available product information; the client, designer and installer remain responsible for the final retail acceptance specification.

• Priority merchandise, skin and brand colours named.
• Exact source/variant and all available metrics recorded.
• Missing Ra, R9, TM-30 or SPD fields marked unknown.
• Equal geometry, illuminance, surroundings and operating state.
• Visual observations separated from measured or supplier-stated data.
• Pass, hold, sample again or request evidence decision retained.