Will you publish our poll results?

No. Commissioned research belongs to the client and is confidential. We also do not publish election projections of any kind.

How do you handle respondents who hide their true preference?

We use indirect questioning, neighbourhood-perception items and turnout-propensity screens, and we report the resulting uncertainty rather than presenting a false precision.

Opinion Polling & Surveys

Statistically sound measurement of where the seat actually stands.

Baseline, tracking and post-campaign surveys designed for Indian constituency realities — stratified sampling, trained enumerators and disclosed methodology.

Book a Strategy Call All Strategic Intelligence

Short answer

What is Opinion Polling & Surveys?

Baseline, tracking and post-campaign surveys designed for Indian constituency realities — stratified sampling, trained enumerators and disclosed methodology.

The Challenge

What usually goes wrong

Poor polling is worse than no polling. Convenience samples, leading questions and enumerators who record what pleases the client produce numbers that feel reassuring and steer the campaign into a wall.

Our Approach

How we run it

Sample design

Stratified random sampling proportionate to booth-level population, weighted for turnout propensity rather than raw population.

Instrument design

Questionnaires are written to avoid leading phrasing and are field-tested in the local language before full deployment.

Enumerator training and audit

Every enumerator is trained and back-checked; a fixed percentage of interviews are independently re-contacted for verification.

Analysis and weighting

Responses are weighted to known demographic and turnout benchmarks, with confidence intervals reported openly.

Tracking cadence

Repeat waves at agreed intervals show movement over time — the number that matters far more than any single snapshot.

Deliverables

Baseline survey with full cross-tabulations
Tracking waves at agreed intervals
Voter sentiment and issue movement analysis
Candidate favourability and recall metrics
Methodology disclosure document

Tools & Data Sources

Stratified sampling frame from electoral rolls
CAPI tablet-based data collection
Back-check and audit protocol
Statistical weighting model

Engagement Timeline

What happens, and when

Week 1
Instrument design and pilot
Week 2–3
Field deployment
Week 3
Audit and back-check
Week 4
Analysis and reporting
Handover and review

In Practice

Where this was deployed

Assembly 2021Anonymised

First-Time Candidate Breakthrough

Built name recognition from near zero by owning one issue completely.

First-Time Candidate Breakthrough

Recognition is the binding constraint for first-timers, and it cannot be bought quickly. Spreading a limited budget across the full agenda would have produced recognition of nothing in particular.

Read the full case study

FAQ

Questions we are asked

Typically 1,200–2,000 completed interviews for a standard Assembly seat, which supports segment-level analysis rather than a single topline number.

What usually goes wrong

How we run it

Sample design

Instrument design

Enumerator training and audit

Analysis and weighting

Tracking cadence

Deliverables

Tools & Data Sources

What happens, and when

Where this was deployed

First-Time Candidate Breakthrough

First-Time Candidate Breakthrough

Questions we are asked

What sample size do you recommend for an Assembly constituency?

Will you publish our poll results?

How do you handle respondents who hide their true preference?

Services that pair with this