NCCS Classification in India: Categories, the Grid and How to Use It

NCCS is a household classification, not a person classification. A household gets one grade, and every member inherits that grade for reporting. It is not an income measure and was never designed to report rupees.

The grid uses education and durable ownership as proxies for affluence, because direct income questions are unreliable in Indian fieldwork. It is quick to ask and cheap to code, which is why it spread through media and FMCG research.

NCCS is separate from the older SEC system. SEC combined the occupation and education of the chief wage earner. NCCS dropped occupation and added consumer durables.

Input one is the education of the chief wage earner, meaning the person who contributes the largest share of household income. Ask for the highest level completed, not the highest level started.

Input two is ownership of eleven listed items. MRSI's 2024 ISEC deck names them exactly:

• Electricity connection
• Ceiling fan
• LPG stove
• Two-wheeler
• Colour TV
• Refrigerator
• Washing machine
• PC or laptop
• Four-wheeler
• Air conditioner
• Agricultural land

Count at the household level, not the respondent's personal use. A two-wheeler used mainly by a son still counts for the household.

Do not treat a cooler as an air conditioner. The list names the air conditioner, and the two items sit at different price points.

The two inputs place a household in one cell of the grid. MRSI labels the twelve cells A1, A2, A3, B1, B2, C1, C2, D1, D2, E1, E2 and E3.

MRSI's ISEC deck reports this IRS 2019 distribution:

Read the shape before you set quotas. C1 and C2 together hold close to three households in ten, so a quota that over-recruits A1 will miss most of the market.

The exact cut-offs that turn an education level and an item count into a cell are published by MRSI in its classification grid. Do not rebuild them from memory or from a blog table.

Durable ownership rose across India, so the same items stopped separating households. MRSI's deck compares NFHS rounds and reports electricity in 97% of households, clean cooking fuel in 59%, and unshared toilets in 69%.

The effect on the grid is direct. When nearly every household owns the base items, the count stops moving households apart. MRSI reports that 93% of shoppers classified NCCS E in 2014 were in C or D by 2022, while their buying behaviour barely changed.

MRSI introduced ISEC in 2024. It uses the occupation of the chief earner together with the education of the highest educated male adult and the highest educated female adult in the household.

ISEC 1 to 6 is the upper group, 7 to 8 upper middle, 9 to 10 middle, 11 lower middle and 12 lower.

If a brief says NCCS, deliver NCCS and state which system you used. Mixing NCCS and ISEC in one report breaks comparability, even when both produce twelve grades.

Ask two questions, not one. First, the highest level of education completed by the chief wage earner, using options that match the grid's bands. Second, a checklist of the eleven items with a simple owned or not owned choice.

Do not ask for a composite score, and do not let an interviewer assign the grade on the spot. Collect the inputs and derive the grade afterwards, so the classification can be recalculated when the grid is updated.

Record the respondent's relationship to the chief wage earner. A student answering for a household is reporting someone else's education and income.

For online panels, keep the checklist tight and unambiguous. Put the eleven items on one screen and require an answer for each, so a skipped item is not read as not owned.

Take a household where the chief wage earner completed higher secondary. The home owns a ceiling fan, an LPG stove, a colour TV, a refrigerator and a two-wheeler. It has no four-wheeler, air conditioner, laptop or agricultural land, and the electricity connection is shared.

The inputs are a higher-secondary education and five of the eleven items. Those two values lead to one cell in MRSI's grid. Read the grade from the published grid rather than estimating it from the count alone.

The point of the example is the process: two clean inputs, one published lookup, one grade. That is repeatable across interviewers and across waves, which is why the system lasted as long as it did.

Set quotas on the grades you will actually report. If the analysis only compares top against bottom, two quotas are enough. If you report six cells, each cell needs enough completes to survive subgroup analysis.

Show the achieved share beside the IRS 2019 shape, so a reader can see whether the sample over-recruits the upper grades.

Label the system and the year. An NCCS grade collected now is not comparable with an ISEC score, and a re-interview can classify the same household differently as durables are added.

For rural samples, ask about agricultural land on its own. It is one of the eleven items and it is easy for an interviewer to skip.

Add NCCS as a screening field, set a quota per grade, and let the platform close a cell when it fills. The education question and the eleven-item checklist fit in one screener.

Test the wording on the free 30 responses before paid fieldwork, then check the achieved grade mix against the distribution above before you report it.