GurjasEvidence & Policy Analytics
Capabilities

Four priority offers. One standard of proof.

Start with the buyer-specific outcome you need. Each priority offer makes fit, deliverables, inputs, timing, indicative investment and ethical boundaries visible before scoping.

Published open evidence

From analytical architecture to decision-ready evidence.

Two reproducible research assets show the scale, provenance and analytical pathways behind our impact-analytics work. Every figure below links to its public deposit.

Open dataset · Version 1.0 · CC BY 4.0

Global Findex resilience analysis

Individual- and country-level conversion measures built from named official sources, with a complete Python pipeline for multilevel, panel, heterogeneity and configurational analysis.

Adults
101,528
Economies
97
Panel
2021–2024
Open dataset, codebook and pipeline DOI: 10.5281/zenodo.20932393
Global resilience analytical architecture A flow from a sample of 101,528 adults in 97 economies through four complementary analytical methods to triangulated evidence. ANALYTICAL SAMPLE 101,528 adults 97 economies Official open sources Multilevel models Country panel Random slopes fsQCA · NCA TRIANGULATED EVIDENCE
Analytical pathways disclosed in the public replication deposit—not a claim of client impact.
Open index · Version 0.1 · CC BY 4.0

AUCR DPI–Inclusion Conversion Index

A reproducible framework measuring how Indian states and union territories convert financial access into use, capability and resilience across three weighting schemes.

States/UTs
36
Analytical sample
34
Scheme agreement
0.89–0.94
Open index, codebook and pipeline DOI: 10.5281/zenodo.20860992
AUCR conversion framework and robustness range Four connected pillars—Access, Use, Capability and Resilience—with agreement across equal, entropy and PCA weighting schemes ranging from 0.89 to 0.94. ACCESS → USE → CAPABILITY → RESILIENCE ACCESS USE CAPABILITY RESILIENCE WEIGHTING-SCHEME AGREEMENT · SPEARMAN 0.890.94 0.801.00
Equal, entropy and PCA schemes remain strongly aligned; the displayed range is reported in the public deposit.

Evidence label: published, independently accessible research assets. Dataset scale and analytical design are factual; no programme outcome or client result is implied.

Working practices

The questions serious teams ask first

Software, timelines, ethics approvals and the use of AI — answered upfront, so nothing surfaces mid-engagement.

SOFTWARE & PACKAGES

Your stack, not ours

We work in R, Python, Stata, SPSS/AMOS and SmartPLS, and deliver in the package your team already uses. If your researchers are committed to R, the entire engagement — code, outputs, replication files — is delivered in R.

TIMELINES

Your deadline sets the clock

Scoping starts by understanding your timeline, then we respond with what is realistically achievable. As a guide: enquiries answered within two working days; a methodology consult typically runs 1–3 weeks; a full evaluation 6–12 weeks. Indicative only — the agreed timeline is fixed in the written scope.

ETHICS & IRB

Approvals before fieldwork

For studies involving primary data from human participants, we ask about institutional ethics / IRB approval at scoping — and can support preparing the application. Where approval is required, fieldwork does not begin without it.

USE OF AI

Human-led, disclosed, never substitutive

AI tools may assist with bounded tasks — literature scanning, code drafting and editing, formatting. They never generate findings, analysis, citations or academic text for a client to submit as their own; every substantive output is produced and verified by the team, and client data is never entered into public AI systems. See our full AI transparency statement.

How engagements run

Five stages, no surprises

A controlled commercial pathway from first diagnostic to final handover. Select a stage to see what enters, what is agreed and what leaves it.

These five stages govern the commercial side of an engagement; the research itself follows the four-phase engagement process described on our homepage. Common questions about pricing logic, authorship and turnaround are answered plainly on the FAQ page.

Get started

Start with a clear research problem.

Use our structured intake form — objective, data status, timeline and expected output — and we respond with a suitable scope, process and next steps, usually within two working days.