When a light fixture is specified, most attention goes to brightness, beam angle, and color temperature. One number that is frequently overlooked — and that has an outsized effect on how a space actually looks — is the color rendering index, or CRI. It determines whether the colors of materials, merchandise, food, skin, and artwork appear as they genuinely are, or whether they appear shifted, muted, or subtly wrong in ways that are immediately felt even when not consciously identified.

Understanding what CRI measures, how the scale works, and where it matters most is foundational knowledge for anyone involved in specifying or evaluating light fixtures.

What CRI measures

The color rendering index is a quantitative measure of how accurately a light source reveals the colors of objects it illuminates, relative to a reference light source. The reference used depends on the color temperature of the source being measured: for lamps below approximately 5000K, the reference is a blackbody radiator (similar to incandescent light); for lamps at or above 5000K, the reference is daylight.

The scale runs from 0 to 100. A score of 100 means the light source renders colors identically to the reference — in other words, colors look exactly as they would under a tungsten bulb or natural daylight. Lower scores indicate that one or more colors are shifted or suppressed relative to how they appear under reference conditions.

CRI is calculated by measuring how accurately a light source reproduces a standardized set of test color samples, known as R1 through R15. The general CRI figure — the one printed on most product data sheets — averages the results of R1 through R8. These eight samples cover a range of moderate saturation colors. The extended R9 to R15 samples cover more saturated and specialized colors, including saturated red (R9), skin tones (R13), and leaf green (R12).

The perceptual impact of low versus high CRI

The difference between a CRI 80 and a CRI 95 source is not always dramatic in isolation. Where it becomes immediately apparent is when the two are placed side by side, or when objects with complex, saturated colors are introduced. A piece of clothing that appears warm burgundy under a high-CRI source may read as flat brown under a CRI 80 lamp. Fresh food that looks vibrant and appetizing under CRI 90 illumination can look pallid and less appealing under CRI 75.

The perceptual effect is experienced most acutely in three ways. First, saturation: colors under low-CRI light appear flatter, less saturated, and less distinct from one another. Second, hue accuracy: individual colors can shift — reds may appear more orange, blues more gray, greens more yellow-green — in ways that misrepresent the actual color of an object. Third, detail: because color is part of how the eye distinguishes between surfaces, low-CRI lighting reduces the apparent texture and depth of materials.

Why the R9 value matters separately

The general CRI figure, averaged from R1 to R8, does not include saturated red (R9). This is significant because saturated red is among the most demanding colors for a light source to reproduce accurately, and it is among the most perceptually important — skin tones, food, textiles, artwork, and flowers all contain substantial red components.

A lamp can achieve a CRI of 90 while having a very low R9 value — sometimes close to zero or even negative (which is mathematically possible under the standard). This means the general CRI figure alone can be misleading in applications where red rendering is important. When evaluating fixtures for retail, hospitality, food display, or any environment where skin tones must appear natural, R9 should be requested and evaluated alongside the headline CRI number. A high-quality source for these applications will typically have an R9 value of 50 or above; values above 80 are considered excellent.

Applications where high CRI is essential

CRI and color temperature: two independent variables

Color temperature and CRI are frequently confused, but they measure entirely different things. Color temperature describes how warm or cool a light source appears — its position on the amber-to-blue-white spectrum. CRI describes how faithfully it renders the colors of objects it illuminates.

The two properties are independent. A 2700K warm-white lamp can have a CRI of 80 or a CRI of 97. A 4000K neutral-white lamp can equally be high or low CRI. Both must be specified independently. A common mistake in fixture specification is to select a correct color temperature and assume that color rendering will be adequate — CRI must be confirmed separately from the product data sheet.

CRI in LED sources: specific considerations

Early LED technology had a well-documented limitation with CRI, particularly in the rendering of warm reds and skin tones. Subsequent advances in phosphor formulation and chip technology have largely addressed this, but performance varies significantly between manufacturers and product grades. A CRI value on a data sheet reflects the performance of the LED package and phosphor combination, not the fixture category — two physically identical luminaires from different manufacturers can have meaningfully different CRI values.

For applications where CRI matters, it is worth requesting spectral power distribution (SPD) data in addition to the headline CRI figure. An SPD graph shows the relative output of a source across the visible spectrum and reveals weaknesses — particularly gaps in the red end of the spectrum — that the averaged CRI number may not fully capture. Sources with a smooth, continuous SPD generally perform better on R9 and on perceptual color quality than those with pronounced peaks and gaps, even when the general CRI figures are similar.

The TM-30 standard: a more complete picture

The CRI standard has been in use for decades, and its limitations are acknowledged within the lighting industry. The most significant is the small sample set (R1–R8) used to calculate the general figure, and the exclusion of saturated colors from the average. In response, the Illuminating Engineering Society developed TM-30, a more comprehensive color evaluation method that uses 99 reference color samples and reports two separate metrics: Rf (fidelity, roughly analogous to CRI) and Rg (gamut, which measures whether the light source makes colors appear slightly more or less saturated than the reference).

TM-30 provides a more nuanced picture of color rendering performance, particularly for sources at the higher end of the CRI scale where differences in quality are not easily captured by the general CRI figure alone. It is increasingly included in technical datasheets from leading manufacturers and is worth referencing when specifying for demanding applications.

Practical specification thresholds

Recommended CRI minimums by application

Energy efficiency and CRI: the trade-off

Higher CRI generally comes at some efficiency cost. Producing a broader, more continuous spectrum — which is what enables accurate color rendering — requires a different phosphor profile than a source optimized purely for lumen output per watt. In practical terms, a CRI 95 LED source may be 10 to 20 percent less efficient (in lumens per watt) than a CRI 80 source from the same product family.

In most applications where high CRI is specified, this trade-off is worth accepting. The purpose of lighting in a retail or residential interior is not purely photometric — it is perceptual. A slightly less efficient source that accurately represents the colors of the space delivers more of the intended function than a higher-efficiency source that distorts them. The efficiency gap between high-CRI LED sources and general-CRI sources has also narrowed considerably as LED technology has matured.

How to read a CRI figure on a data sheet

Product data sheets typically present CRI as a single number, often labeled "CRI," "Ra," or "CRI (Ra)." The Ra notation indicates that the figure is the average of R1 through R8. This is the general CRI. R9 is often listed separately, if at all — and its absence from a data sheet is itself informative. Manufacturers confident in their R9 performance tend to include it; those with weak red rendering often do not.

When comparing two fixtures with the same headline CRI, request full R1–R15 data or SPD curves. The shape of that data — particularly the values for R9, R12, and R13 — will reveal differences in performance that the averaged figure obscures. This level of detail is available from reputable manufacturers on request and should be standard practice when specifying for demanding environments.

Consistency across a lighting scheme

In a layered lighting scheme — where ambient, task, and accent sources from different fixture types are used together — CRI consistency across all sources matters. If the ambient layer is specified at CRI 80 and the accent layer at CRI 95, the colors of objects in accented areas will appear noticeably different from those in the ambient field, creating a visual inconsistency that is difficult to diagnose without checking the specifications. Where color rendering matters to the overall effect, all circuits in the same visual field should be specified to the same CRI standard.