All articles
10 min read

Packaging Research in India: What to Test and How Many People

Packaging research in India works best in two steps: interview 8 to 12 shoppers per segment to find confusion, then survey 200 to 390 people per design only if you must choose.

By Jitendra Kumar Nirala, Founder · Updated 6 Nov 2026

A researcher with a clipboard listens as a woman shopper in an Indian grocery store examines a packaged snack box.

Key takeaways

  • Run packaging research in two steps: interviews and observation to fix what people misread or cannot find, then a survey only if you still have to pick between designs.
  • Plan 8 to 12 interviews per audience segment to find problems. In a survey, about 200 per design is where a 10-point gap in your sample becomes statistically significant. About 390 per design is needed to detect a true 10-point gap with 80% power.
  • A preference survey with 50 people per design only shows gaps of about 20 points or more as significant. Gaps between similar finished designs are often smaller than this, so a 50-person group is better used to find problems than to pick a winner.
  • Food and drink brands should check that shoppers can explain a front-of-pack warning in their own words. As of 6 November 2026, the Supreme Court has reserved its order on front-of-pack warning labels (hearing of 28 September 2026) and has not yet pronounced it.
In this article
  1. 1What is packaging research, and which kind do you need?
  2. 2How many people does packaging research need?
  3. 3Why do pack tests so often answer the wrong question?
  4. 4What should a packaging research brief contain?
  5. 5How do we run a pack test in India?
  6. 6What should food and drink packs check before front-of-pack warnings arrive?
  7. 7Frequently asked questions

A 50-shopper test of two packs cannot tell a narrow win from a coin toss. The working is further down this page, and it changes how a pack test should be planned.

Packaging research often mixes up two jobs. One is finding out whether a pack is understood, found and usable. The other is choosing between finished designs. The first needs a few dozen conversations. The second needs hundreds of survey responses. If you pay for the second before doing the first, you get a confident winner among packs that still confuse people.

This page covers the consumer side of the subject. Journals on barrier films, materials and transport testing are laboratory work and a different field. Below you will find a question-to-method table, a sample-size table with its formula, a brief checklist, and a note for food packs facing front-of-pack warning labels.

What is packaging research, and which kind do you need?

Packaging research asks real shoppers four things about a pack: whether they understand it, find it, can use it, and prefer it to the alternatives. Each question needs a different method, so decide which one you are asking before you book anything.

Your questionMethodWho and how many
Does the pack say what it is and who it is for?In-depth interview (IDI, a one-to-one conversation of 30 to 120 minutes) or focus group (FGD, a moderated discussion of 1 to 3 hours)8 to 12 per segment
Can shoppers find the right variant among competitors?Shop-along: you observe a shopper in a store or on a shopping app8 to 12 per segment
Can people open, pour, store and reseal it?In-home use test (IHUT) or usability session5 to 8 per user group
Do people understand the claim or the label panel?IDI or online bulletin board (a discussion held over several days, not live)8 to 12 per segment
Which of 2 to 4 finished designs wins?Survey where each person sees one design only (a monadic test)About 200 or more per design

A segment is a group you want to hear from separately, such as one city and one type of shopper. In a survey, a cell is the group of people who see the same design.

An in-home use test has participants use the real product at home, which is where opening, storage and reuse problems show up. A shop-along shows what a shopper actually does at the shelf, which a slide of your pack cannot. Studying people in their own setting is what the Nielsen Norman Group, a user research consultancy, calls a field study: research that takes place in the user's natural environment (published 12 January 2024).

How many people does packaging research need?

Plan 8 to 12 interviews per audience segment to find problems with a pack. In a survey, about 200 per design is where a 10-point gap first shows as significant, and about 390 per design is needed to detect a true 10-point gap reliably. Smaller surveys only show much larger gaps.

For finding problems. For a typical qualitative usability study of a single user group, the Nielsen Norman Group recommends five participants to uncover the majority of the most common problems (published 1 December 2019, last reviewed 15 July 2026). For packs, our recommendation is 8 to 12 per segment. This is our recommendation, not a published standard: five per group finds most common problems, and we add three to seven because shoppers differ by city and language. Two cities and two shopper groups make four segments. At 8 each, that is 32 interviews. For more on sizing qualitative studies, see sample size for qualitative research.

For choosing a winner. The number depends on how big a gap you need to see. This table is our own calculation, so you can redo it. It gives the smallest gap between two packs' preference shares (the share of people preferring each pack) that shows as statistically significant, using the formula gap = 1.96 × √(2 × 0.25 ÷ n), where n is respondents per design. The 7.4-point figure for 350 per design comes from the same formula and inputs.

A researcher and a young man sit across a small table with one product pack between them during an interview.
Small one-to-one interviews are enough to surface most confusion with a pack.
Respondents per designSmallest gap that shows as significant (percentage points)
3025.3
5019.6
10013.9
2009.8
3507.4

The assumptions are two designs, each shown to different people, shares near 50%, 95% confidence, a two-sided test (it checks for a difference in either direction) and equal group sizes. 95% confidence means that if there were no real difference, a gap this large would appear by chance in only about 1 test in 20. Real samples are rarely this clean, so treat the table as a minimum.

A 55 to 45 result is a 10-point gap. With 50 people per design, a gap that size can easily be chance. With 200 per design it just clears the line, but a true 10-point gap would show as significant only about half the time. To detect a true 10-point gap with 80% power (a real gap that size would show as significant in about 8 tests out of 10) at 95% confidence, you need about 390 per design. This is our calculation: n = 2 × (1.96 + 0.84)² × 0.25 ÷ 0.10².

What a survey costs. The cost is cells × responses per cell × the rate per response a provider quotes. With an illustrative rate of ₹X per response, the number of responses you pay for looks like this:

Designs200 each350 each
2400 responses, 400 × ₹X700 responses, 700 × ₹X
3600 responses, 600 × ₹X1,050 responses, 1,050 × ₹X
4800 responses, 800 × ₹X1,400 responses, 1,400 × ₹X

We price each project separately, in a tailored quote.

Why do pack tests so often answer the wrong question?

A common mistake is to ask "which do you prefer?" before checking whether people understand the pack, and then to treat small gaps as winners. A pack can be liked and still misread. A preference score will not show you that.

There are three common mistakes.

  1. 1
    Asking about liking first.
    A shopper can say a pack looks premium and still believe it holds the wrong product. Show the pack briefly, take it away, then ask what the product is, who it is for and what they noticed first. Do this before any question about preference.
  2. 2
    Reading a small sample as a verdict.
    The sample-size table above shows how large a gap must be before it counts.
  3. 3
    Testing the pack on a slide.
    The same pack is seen differently on a kirana counter (a neighbourhood shop), on a supermarket shelf and as a small thumbnail on a quick-commerce app (one that delivers in minutes). Test it where it will be sold. In our view, name one main setting in your brief and test there first.

What should a packaging research brief contain?

A packaging research brief should state the decision, the pack versions, the competitors, the setting, the languages, the cities, the claims to check and the deliverables. Each item below is a test you can run on your own brief. Our one-page market research brief template gives the wider structure.

  • Decision. Write one sentence: "If X, we will change the pack; if Y, we will print." Pass: someone outside your team can say what result would change the pack. Fail: the brief says "get feedback".
  • Pack versions. Pass: each version is a print-quality mock-up or the physical pack at real size. Fail: a slide image.
  • Competitor set. Pass: you name the packs shoppers see beside yours in the chosen setting. Aim for 3 to 5. Fail: "similar products".
  • Setting. Pass: one main setting is named (kirana, supermarket or quick-commerce app). Fail: "retail".
  • Languages. Pass: every language printed on the pack is read by at least some participants who read it daily. Fail: only English testers for a bilingual pack.
  • Cities. Pass: they include the cities where first distribution happens. Fail: cities chosen by convenience.
  • Claims. Pass: for each claim you list its intended meaning in one sentence, so you can compare it with what participants say. Fail: "check the claims".
  • Deliverables. Pass: recordings, transcripts, data exports or a report are agreed before the study starts. Fail: "findings".

How do we run a pack test in India?

We run a pack test as a recruitment-and-fieldwork project built from your brief. You send the brief; we check feasibility and send a tailored proposal and quote. The qualitative formats we field include IDIs, FGDs, shop-alongs, in-home product tests, usability and app testing, and online bulletin boards.

A fieldworker with a tablet watches a woman open a boxed food product at home.
Fieldwork puts the pack in real hands at home, where shoppers actually use it.
  1. 1
    Brief.
    You share the objective, method, target audience, geography, sample size, timelines and recruitment criteria.
  2. 2
    Proposal.

    We assess feasibility and send a commercial quote. We price each project separately, because method, audience, geography and how hard the people are to recruit all change the cost. A free pilot or sample is available; ask when you write. For a sense of how line items add up for one format, see our focus group cost checker.

  3. 3
    Recruitment.
    We recruit by city, target by household and demographic profile, run participant identity checks before invitation, and can run sessions in languages other than English.
  4. 4
    Fieldwork.
    Participants are scheduled for the agreed sessions. Their incentives are processed after participation is completed and validated.
  5. 5
    Deliverables.
    Depending on scope, you receive recordings, transcripts, structured data exports and a research report, all agreed before the study begins.

We can recruit and manage participants for both the interview step and a survey step, with pricing set per project in a tailored quote.

To start, email your brief to support@qualfacto.io, or message us on WhatsApp for quick questions. We reply within one business day where we can, Monday to Friday, 10:00 AM to 6:00 PM IST. You can also read about our qualitative research panel. If you want to take part in studies like these, register on how to join and get invited; registration does not guarantee selection, and invitations depend on each study's requirements and your profile.

What should food and drink packs check before front-of-pack warnings arrive?

Food and drink brands should check whether shoppers can say, in their own words, what a front-of-pack warning means and which nutrient triggered it. The proposed warning label is not yet final, so test comprehension of whatever panel your pack may carry.

As of 6 November 2026, the Supreme Court has reserved its order on front-of-pack warning labels for packaged food and has not yet pronounced it. At the hearing of 28 September 2026 it questioned the proposed timeline of the Food Safety and Standards Authority of India (FSSAI), as reported by the legal news outlet LiveLaw. Its proposal, as reported, is a red hexagon on a white square. The warning appears when any one of added sugar, added fat or salt crosses a threshold. FSSAI proposed about four months to finalise the draft plus a 365-day voluntary period, and the court questioned that. The details may change with the order.

In interviews, show the pack with a mock-up warning and ask the shopper what it means. Our recommended pass mark is at least 7 of 8 participants per segment naming the nutrient correctly without prompting. Anything lower means the panel, or what the eye sees first around it on your pack, needs work.

Send us your pack brief for a tailored quote

Email your brief to support@qualfacto.io and we will check feasibility and send a tailored proposal and quote.

Send your brief

Frequently asked questions

Packaging research is the study of how real shoppers understand, find, use and choose a pack. Material and transport testing is a separate laboratory field. This page covers the consumer side only.
Package testing in the engineering sense measures things like shock, vibration, compression and barrier performance in a laboratory. Packaging research asks consumers what they understand and do. A pack can pass every lab test and still confuse shoppers.
It depends on the gap you want to see. By our formula, 350 per design shows a gap of about 7 points as significant, and 200 shows about 10. To detect a true 10-point gap reliably, about 390 per design is needed.
A focus group can show why a pack confuses people, but it should not name a winner. The numbers are small, and in a group one strong voice can sway others. Use it to fix problems, then survey if you must choose.
Start with open questions that let shoppers describe the pack, its buyer and its main claim in their own words. Put rating questions, such as liking and purchase intent, last.

Sources

  1. 1.Our calculation of the smallest gap between two preference shares that shows as significant, using gap = 1.96 × √(2 × 0.25 ÷ n). It assumes shares near 50%, 95% confidence, a two-sided test and equal cells. Our power calculation for a true 10-point gap uses n = 2 × (1.96 + 0.84)² × 0.25 ÷ 0.10², which gives about 390 per design. The cost table multiplies cells × responses per cell × an illustrative rate of ₹X.
  2. 2.Usability Testing 101, Nielsen Norman Group, published 1 December 2019, last reviewed 15 July 2026
  3. 3.Field Studies, Nielsen Norman Group, 12 January 2024
  4. 4.Supreme Court asks FSSAI why one year is needed for front-of-pack labels and reserves judgment, LiveLaw, September 2026

Keep reading