B2B Decision Maker Survey in India: How to Design It and Where It Gets Hard

Tie the study to one recent purchase decision in a defined category, such as accounting software, packaging material or point-of-sale hardware. 'B2B in general' produces answers that fit no one.

Most business purchases involve more than one person. The roles below are a common way to describe them. Ask each respondent which role or roles they played in the last purchase, and allow more than one.

Add a question on the size of the group: how many people took part in the decision. Use bands, not an open number, and add 'I do not know'. Compare the answers by role. A user and a decider often describe different decisions, so report them separately.

For stages, list the steps in plain words: noticed the need, collected information, compared vendors, negotiated, approved, signed, implemented. For each, ask whether they were involved and who else was. Add one open question: 'What nearly stopped this purchase?'

Build the vendor funnel with carry-forward routing. Show a list of vendors and ask which they have heard of. Carry those into 'which would you consider', carry those into 'which did you shortlist', and then 'which did you buy from last'. Our survey builder documents carry-forward and option loops, so each stage narrows the list automatically. See our skip logic guide. The drop between two stages tells you where the vendor loses accounts.

Randomise the order of the vendor list so no name gains from sitting first, and include a vendor that does not exist to catch over-claiming.

Buyers say price, quality, service, support and delivery all matter. A rating grid will confirm it. Use MaxDiff instead: show small sets of criteria and ask for the most and least important in each. It works for roughly 8 to 30 items and forces a trade-off.

Write criteria in concrete terms, such as 'delivery on the date promised' or 'a named person to call when something fails', not 'good service'. For a feature or claim list, the same design is covered in the MaxDiff guide.

For vendor NPS, use the standard 0 to 10 likelihood-to-recommend question and ask it once per vendor the respondent currently uses, as an option loop over the vendors they picked. NPS is the percentage scoring 9 or 10 minus the percentage scoring 0 to 6. See NPS methodology for the cautions on small bases.

Screening decides whether the study is valid. Put the screener first and build it like this.

Our survey builder documents screen-outs, complete-early routing and consistency checks. It also documents screener integrity: category questions asked neutrally among decoys, with key criteria re-verified in other words. Use all of them.

All of these answers are self-reported. Verification on our panel establishes that a member is a real, unique individual. It does not verify an employer or a job title.

The SuperJ panel is a consumer panel of members of the public in India. It includes professionals, such as doctors and carpenters, but it was not recruited as a list of purchasing heads or company directors. We do not claim a B2B panel and we do not publish counts of any job role.

So the question is incidence: of the members you can target, how many will pass your screener? A broad B2B audience, such as owners of small businesses buying everyday supplies, may be reachable. A narrow one, such as chief information officers at large firms, may not be.

Ask before you commit budget. Give the team your screener, the target sample and the sectors, and ask whether they can give a feasibility count. Screened-out respondents do not count towards your response bundle.

A rough guide to sample: with 100 decision makers, a result near 50% carries a margin of about 9.8 percentage points at 95% confidence. With 150 it is about 8.0 points. These are simple-random-sample figures, an approximation for quota panels. Treat a B2B study with these bases as directional and avoid fine cuts.

If the audience cannot be reached, switch the method. Account lists, telephone interviews or moderated web interviews are better for narrow roles.

If your job is a decision maker survey where the roles are common among members of the public in India, this is what a 150-response study costs at list price. Confirm feasibility first.

Pro is sold in bundles of 400 to 1,000 responses and costs 20% less per response than Plus. Enterprise plans come with a dedicated research team and are priced on request, so raise a narrow audience with them. Targeting covers city, tier, age band, gender (male or female) and NCCS class.

SuperJ has a 25M+ consumer footprint in India only. The survey-eligible panel is a subset, so availability depends on targeting. Members complete Aadhaar-based KYC once, and each study gets a zero-knowledge proof, not identity documents. Speeders, straight-lining and inconsistent answers are removed before analysis. See panel methodology and pricing. The limit is the section above: this is a consumer panel.