How we work.
Sound decisions depend on evidence that holds up to scrutiny. This is how we design, collect, validate and analyse every dataset, so each figure we report can be traced to its source and defended.
Five stages, each with its own checks
1 · Research Design and Sampling
Every study starts from the decision it has to inform. We define the target population, build the sampling frame from the most current sources available, and set stratification and quotas by region, urban or rural location, age, gender and income band, so the sample reflects the market you are measuring. Sample sizes are set for the precision the decision needs, and every assumption is documented before fieldwork begins.
2 · Data Collection
Trained enumerators interview in person across Nigeria's 36 states and the FCT, using tablet questionnaires that record GPS location and time on every interview. Where the audience calls for it, we add telephone (CATI) and online panel interviews in a mixed-mode design. Field supervisors track completion rates, interview length and quota progress daily, and correct problems while the team is still in the field.
3 · Quality Assurance
Every response passes automated and manual checks before analysis. We verify location and timing data, remove duplicates, and back-check a share of interviews by phone or revisit. Speeding, straight-lining, logical inconsistencies and implausible outliers are flagged and investigated, and records that fail are removed or replaced, never quietly corrected.
4 · Statistical Analysis
Our statisticians weight the clean data to known population benchmarks, then apply the method the question calls for: descriptive estimates, significance testing, regression, segmentation or forecasting. Results carry their base sizes and margins of error, and a difference is reported as real only when it is statistically significant.
5 · Reporting and Delivery
Findings lead with the answer, followed by the evidence and the limits of what the data can support. Every project includes an executive summary and a full methodology appendix, and the cleaned dataset and analysis files are available on request, so your team can verify any figure for itself.
Have a question about our methods?
We will gladly walk you through the sampling plan, weighting scheme or quality checks behind any study. Methodological transparency is part of every engagement.
Talk to our team