OBSERVATION
What a source measured
Retain the population, unit, geography and observation period.
THE RESEARCH STANDARD
Useful research.
Visible assumptions.
How the atlas turns public evidence into questions worth exploring, and where its conclusions stop.
Full-length, branded PDFs. Enter your email to view or download.
THE RESEARCH METHOD / AN OPEN PROCESS
From a source to a better question.
The atlas makes the route from evidence to interpretation visible. Each step has a different standard of proof.
- 01Source
Find public, traceable evidence
Official statistics, city plans, regulators and first-party programme documentation.
- 02Scope
Attach geography and time
Retain the source unit, observation year, target year and uncertainty.
- 03Question
State an editorial hypothesis
Identify a possible buyer, useful service and tradeoff.
- 04Test
Design local validation
Agree a baseline, comparison, safeguard and outcome before making a claim.
New evidence can change the question at any point.
A desk-research workflow. Local interviews, customer commitments and pilot outcomes remain work to do, rather than evidence already collected.
THE CLAIM STANDARD / KEEP THE CATEGORIES CLEAR
A number is not yet an outcome.
Read each category on its own terms. The atlas does not turn these different forms of evidence into one score.
TARGET / PROJECTION
What is intended or expected
Keep the future date and uncertainty visible. Achievement still needs evidence.
EDITORIAL HYPOTHESIS
What may be worth testing
A proposed buyer, pilot or business model requires local discovery.
VERIFIED OUTCOME
What a test can establish
Requires a baseline, comparison, costs and safeguards. No such field result is claimed here.
A guide to interpretation used throughout the atlas. Status labels preserve what the underlying source can actually establish.
Start with the place.
The atlas covers the 12 cities already featured on the Pheonix map. They are a defined editorial sample, not a ranking, an exhaustive global survey, or a statement that a chapter is operating there. City boundaries, metropolitan regions, states and countries are different units; each indicator retains the geography reported by its source.
Use evidence readers can follow.
We prioritise official statistics, city plans, central banks, regulators, multilateral institutions and first-party programme documentation. Every brief links to the underlying sources. Government plans establish stated policy intent; programme websites establish what an institution says it does. Neither proves commercial demand, implementation success or independent evaluation.
Separate data from the date of publication.
A 2026 publication may describe a 2023 observation. Signals show data years, policy target years or forecast labels where relevant. This edition was assembled on 14 October 2026. It is a dated desk-research snapshot and does not update automatically. Linked documents may subsequently change.
Treat opportunities as hypotheses.
The business opportunities, potential buyers and pilots are editorial interpretations of the cited context. We have not represented them as customer commitments, field interviews, investable opportunities or verified demand. Pilot durations, sample sizes and economics must be designed locally after discovery.
Make impact falsifiable.
For each proposal, establish a baseline, identify the people affected, define a useful outcome, and test what would have happened otherwise. Measure costs and negative effects as well as benefits. Disaggregate outcomes where appropriate, protect personal data, and involve the people expected to benefit. SDG tags indicate thematic alignment, not UN endorsement or certified SDG performance.
Compare questions, not invented scores.
The comparison view places local hypotheses side by side within the same market. It deliberately avoids aggregating incompatible currencies, policy targets, population estimates or years into a single score. Global investment totals, social need and customer revenue are distinct quantities.
Keep the limits visible.
There is no primary fieldwork, private transaction database, investment model, clinical evaluation or legal opinion behind these briefs. Source quality and geographic detail vary. Programme status, regulation, procurement, tariffs and business conditions need fresh checks before action. The open questions identify what local discovery should resolve next.
Make the work inspectable.
All research is readable without membership. Each brief has a stable URL, source list and a complete, branded PDF for reading and sharing. The full atlas is also available as one PDF. Attribution is Joshua Armah, Founder, Pheonix Startups. Underlying sources retain their own rights; a source citation does not transfer a licence. Corrections can be raised through the Pheonix organiser who shared the atlas.