About Hercules

Hercules Works is a research platform for India. You describe the decision you need to make. The platform’s AI drafts the brief, chooses a research design, writes the questionnaire and its routing logic, and checks it against a quality rubric. The survey then goes to SuperJ members, and the Poseidon analytics engine turns their answers into a report. It is made by Jupiter Meta Labs in Hyderabad, Telangana.

Indian respondents

2 Cr+

Data points generated

5B+

Surveys deployed

2000+

Questions answered

50,000+

The 2 crore figure is the number of Indian respondents on the SuperJ network; the verified, survey-eligible panel is a subset of it.

1. Step 1
   Create
   Describe the study in plain language. AI builds the brief, design, questions and routing. You edit by chat.
2. Step 2
   Field
   Deploy to SuperJ. Target by city, age, gender and NCCS class. Members are verified and anonymous.
3. Step 3
   Analyse
   Poseidon reads the responses, answers questions in plain language and writes the report.

Create a survey with AI

Seven steps run behind a single prompt: a research brief, a methodology framework you can approve or refine, the study flow and routing logic, the questionnaire, a quality check against a nine-dimension rubric, deployment to SuperJ, and the handoff to Poseidon. Weak studies are blocked from publishing until the hard-gate checks pass. See the pipeline step by step.

Edit by conversation

Ask for changes the way you would ask a colleague: “reword Q7 to be simpler”, “show Q28 only the options they picked in Q27”, “add a screener for two-wheeler owners”. The editor changes wording, options, question type, routing or design, and tells you what changed and what it could not do. You can also paste or upload an existing questionnaire and have the design and routing built around it.

Question types

• Single and multi-select, yes or no, rating scales (including NPS 0 to 10) and rating grids.
• Ranking, open text and MaxDiff best-worst exercises.
• Images, video and audio as questions or as answer options, for pack, ad and concept tests.

Routing and logic

• Skip logic, screen-outs, complete-early and display conditions.
• Carry-forward, where options picked in one question become the options in a later one.
• Option loops (“for each brand you use”), piping of earlier answers, and consistency checks.
• Monadic and sequential-monadic rotation, and option randomisation.

22 study types, built on named research designs

Each study type comes with an instrument, a question order and the right design, so the KPI is the standard one and not a home-made variant.

• Brand and communication: brand tracking, ad pre-tests and ad testing, campaign tracking and brand lift.
• Product and pack: concept screening and testing, product testing, packaging, design, name and logo testing.
• Price and priorities: Van Westendorp and Gabor-Granger price studies, MaxDiff, Kano and TURF.
• Customers: usage and attitude, satisfaction, customer experience, churn, segmentation, profiling, shopper path to purchase and market sizing.
• Other audiences: B2B decision-makers, public opinion, employee engagement and usability.

Choose a study type or choose a research design.

Reach the audience you need

SuperJ members include general Indian consumers and professionals such as doctors and carpenters, so a study can be pointed at a very specific audience. You can target by:

• Any Indian city, based on your requirement, with quotas per city or tier.
• Gender (male or female) and age band (18-24, 24-35, 35-45, 45-55, 56-90).
• NCCS class, A1 to C1, and behavioural segments such as two-wheeler owners.
• Interlocking quotas across gender, age and city.

How audience targeting works. Respondents are verified and anonymous; see About SuperJ.

Analytics and reports

When a survey completes, Poseidon ingests the responses and builds a model of the survey: its questions, types, routing and demographics. From there you can:

• Ask questions in plain language, such as “how does purchase intent differ by age band?”.
• Get cross-tabs and significance tests, and themes from open-ended answers in the report.
• Get a long-form report in PDF, HTML or Markdown, with the right measure chosen for each question type: NPS, top-2-box, MaxDiff utilities, TURF, driver analysis, segmentation and more.

Data quality is built in

• Attention checks are inserted automatically, roughly one for every seven questions.
• Speeders, straight-liners and inconsistent answers are flagged and removed before analysis.
• Open-ended answers are scored for gibberish, copy-paste and off-topic text.
• Repeat respondents are excluded within set windows.

Data quality checks in detail and how respondents are verified.

What it does not do

Hercules Works does not run choice-based conjoint or constant-sum allocation, and it does not measure task or reaction times. The panel is in India only, it is app-based so respondents are smartphone users, and it is not a global panel. We say so up front because it helps you pick the right tool.

Pricing

The Free plan includes 30 responses. Plus costs ₹150 per response from 100 responses. Pro costs ₹120 per response in bundles of 400 to 1,000. Enterprise pricing is custom, with a dedicated research team.

Common questions