Companion to the v1.1 Methodology publication. 16 May 2026.
Audience: Afrimintel users, recipients of the v1.0 methodology PDF, and institutional readers evaluating the platform’s discipline.
Editorial responsibility: Nikesh Patel, Honorary Consul of Rwanda in Mauritius. nikesh@afrimintel.com.
Methodology v1.1 publishes corrections to v1.0 across seven distinct layers. The customer-visible effect: 11 of 13 published province Scores move downward, by an average of approximately 1.4 points. One Score moves upward (Kalahari Platform). One holds at the v1.0 published value but with its underlying components corrected. The seven layers, the per-province movements, and the regression checklist are at the v1.1 audit log entry. The methodology document itself is at /afrimintel_methodology.pdf.
The v1.1 publication is a launch moment for the platform’s correction discipline, not a routine update. The Quality Standard publishes a 7-business-day correction SLA for material errors; the v1.1 corrections honour that commitment by surfacing the corrections in public with per-layer attribution rather than absorbing them silently into a vintage refresh.
This companion document is the reader’s orientation to what changed and how to verify it.
The Methodology document publishes a formula:
Score = (MSP × 0.6 + IC × 0.4) × Opportunity Multiplier. In
v1.0, 12 of 13 published Scores did not follow from this formula under
the v1.0 input values. Mismatches ranged from +0.27 (Cape Fold Belt,
near rounding tolerance) to +2.01 (Saharan Metacraton). Every mismatch
was in the same direction — the published Score was higher than the
formula-correct value.
In v1.1, every published Score is computed directly from the formula with no unstated adjustments. Anyone with the MSP, IC, and OM values for a province can reproduce the Score to one decimal place.
This is the most consequential layer. It is the reason v1.1 is published as a launch moment rather than a vintage refresh.
The Opportunity Multiplier ranges from 0.80 to 1.25 and reflects situational factors that MSP and IC don’t capture (capital programmes, security signals, infrastructure changes). Two provinces carry revised OM values at v1.1:
All other OM values are held from v1.0. Per-province OM rationale text is now published in the §07 Table 4 OM Rationale column — a new column at v1.1.
The Opportunity Multiplier trigger framework (§05) is refreshed across four items:
The principle: methodology examples drift faster than the abstractions they illustrate. The audit cadence sits at the dossier level where it is closer to source.
This is a new §04 sub-section. Where Fraser Institute does not cover a country in a province’s IC scope, Afrimintel uses Transparency International’s Corruption Perceptions Index as a proxy. Convention:
methodology_v1_1_section_04_calibration.md sensitivity
analysis.The band’s direction is upward by default: most countries score higher on Fraser (mining-investor perception) than on CPI (generic public-sector corruption). One country in the calibration sample inverts — Burkina Faso, where Fraser 2025 IAI 35.29 [Pending Fraser PDF back-check] is below CPI 2024 41 [Sourced]. The mechanism is structural: mining-extractive-specific factors (the 2022 coup, ongoing Sahel security collapse, mining nationalisation direction) drive the Fraser IAI below the generic CPI score.
v1.1 documents this as a structural-trigger inversion framework. Where structural mining-extractive factors apply, the band direction is downward rather than upward. Currently applied to two CPI-proxy components: Ethiopia (Afar Depression; security deterioration in northern Ethiopia, extractive-policy uncertainty) and Niger (Saharan Metacraton component; 2023 coup, ongoing JNIM/IS-Sahel conflict).
The full calibration and the structural-trigger criteria are documented in §04 of the methodology. The Afrimintel Score itself is published as a point estimate at v1.1; the band is captured in the IC source attribution rather than visualised on the Score. Calibrated band-to-Score propagation is committed to v1.2 development.
v1.0 used Fraser Institute Annual Survey of Mining Companies 2024 figures. v1.1 refreshes to Fraser 2025 (published 26 February 2026; survey conducted August–November 2025; 68 jurisdictions; 14 African countries).
Fraser 2025 IAI values, exact decimal, Pending Fraser PDF back-check (n=10): Botswana 85.99, Morocco 78.97, Zambia 72.84, Tanzania 68.04, Côte d’Ivoire 60.92, Ghana 55.21, South Africa 54 (Derived from rank 57/68; exact pending PDF), Guinea 52.16, Mali 46.58, Burkina Faso 35.29. Each value is corroborated across two or more secondary aggregators citing the primary Fraser numbers; back-check against the Fraser Institute’s own published PDF resolves the Pending state. Target resolution: within 30 days of v1.1 publication.
Fraser 2025 IAI values, rank-confirmed-exact-decimal-pending, Pending Fraser PDF back-check (n=4): DRC ~58 (rank 50), Namibia ~56 (rank 51), Angola ~54.5 (rank ~55), Egypt ~36 (rank 66/68). These values are sourced via secondary aggregators citing primary numbers (Pravda Egypt, mining.com, others per audit-log entry), with the bracket-derivation transparent in the audit-log entry. Both the rank-and-bracket attestation and the exact-decimal resolution are pending Fraser PDF back-check. Target resolution: within 30 days of v1.1 publication.
Confirmed NOT in Fraser 2025 — CPI proxy applied: Ethiopia (was rank 82/82 in 2024, dropped from 2025), Kenya, Mozambique, Sierra Leone, plus Algeria, Niger, Chad, Sudan, Libya (non-Mali Saharan Metacraton components).
On the timing. The Quality Standard publishes a 30-day SLA for vintage refreshes from each Fraser publication. v1.1 publishes outside that window. The cause is recorded in the audit log entry: the worksheet walkthrough that surfaced the §07 algebraic correction (Layer 1) required full apply-step recompute before vintage refresh could be applied to corrected Score formulas. The 30-day SLA is reaffirmed for future refreshes.
Methodology v1.0 §08 stated “Data partner relationships (BGS, Fraser Institute, USGS, AfDB) are structured as data exchange agreements.” This overstated the relationship status for all four institutions simultaneously. None of the four had a formal data exchange agreement with Afrimintel at v1.0 publication. The same overstatement pattern was caught and softened in Competitor Gap §02 (BGS) in April 2026; the systemic version in Methodology §08 was missed by open-ended review until the structured worksheet walkthrough surfaced it.
v1.1 §08 is restructured into two sub-sections:
The §08 restructure cascades to five additional customer-facing surfaces beyond the methodology document. Each cascade is its own audit log sub-entry: the Independence Policy mirror page, the Acquirer Brief (DATASET v1.1 → v1.2), the Category Paper (v1.0 → v1.1 with the “partnership architecture” section rewritten as “source citation architecture”), platform copy across the website, and the API schema (v1.1 → v1.2 with five new Province Object fields).
The geological extent of a province often spans more countries than the IC component scopes to. Congo Craton’s geology spans the DRC plus Congo Republic, Cameroon, CAR, and Gabon; the IC component in v1.0 was DRC-only without saying so explicitly. ANS spans Egypt plus Saudi Arabia, Sudan, and Eritrea; v1.0 was Egypt-only. Saharan Metacraton spans six countries; v1.0 inputs implied uneven weighting without publishing the basis. Afar spans Ethiopia plus Eritrea plus Djibouti; v1.0 was Ethiopia-only.
v1.1 publishes an explicit Scope column in §07 Table 4 stating the IC scope per province. This is the Approach 2 baseline (provinces explicitly scoped to a stated country subset). v1.2 is committed to developing Approach 1 (full multi-sovereign IC across the geological-extent country list with documented per-component weights).
The methodological principle that surfaced this layer is durable: geology knows no political borders, but Investment Climate does. v1.1’s Scope column makes the implicit explicit.
Anyone reading this document should be able to reproduce every published Score in the v1.1 methodology. The verification path:
Step 1. Open the v1.1 Methodology PDF at /afrimintel_methodology.pdf. Find §07 Table 4.
Step 2. For any province, pick the row. Take the
published MSP, IC, and OM values. Compute
(MSP × 0.6 + IC × 0.4) × OM. The result should match the
published Score to one decimal place.
Step 3. To verify the IC value, follow the IC Source column. For Fraser-attributed IC values (Pending Fraser PDF back-check per Methodology §07), reproducibility is currently via the secondary aggregators citing primary numbers — with attribution in the audit-log entry — and will be against the Fraser Institute’s own published PDF once back-check completes (target: within 30 days of v1.1 publication). For CPI-proxy IC values, the TI Corruption Perceptions Index 2024 publishes all African country scores at https://files.transparencycdn.org/images/CPI2024_Map-plus-Index.pdf — these values are Sourced from TI’s primary publication.
Step 4. To verify the per-layer attribution and the regression checklist, read the v1.1 audit log entry. It names every input source, every algebraic check, and every cascading sub-entry.
If any step in this verification fails, contact nikesh@afrimintel.com directly. The Correction Velocity SLA applies.
The province Scores are now reproducible from public data sources. A junior analyst at AfDB, TDB, or Afreximbank can verify any Afrimintel Score in under 30 minutes using the v1.1 methodology document plus Fraser Institute and TI publications. The verification trail is the asset.
For provinces where the underlying methodology has shifted materially (Saharan Metacraton, Mozambique Belt, Arabian-Nubian Shield, Afar Depression), the v1.1 Score and the audit log entry’s per-layer attribution should be substituted for any v1.0 Score values in active investment memos or pipeline tracking.
The 7-business-day Correction Velocity SLA is the operational commitment behind the discipline. Users surfacing data errors via nikesh@afrimintel.com receive correction within the SLA window.
MSP values are held constant from v1.0 to v1.1. The geological assessment underlying the Score has not changed. What has changed is the surrounding apparatus: the IC component, the OM rationale per province, the explicit IC scope, and the published formula consistency.
For deposit-level dossiers, the v1.1 publication does not change deposit-specific data (resource estimates, ownership stakes, technical reports). Those data are governed by the dossier-level audit cadence (quarterly deep audits on rotating provinces; daily watchlist scan; monthly 4-criteria province review) and the Quality Standard three-state model.
The Country Pitch Builder uses Investment Climate values that are now refreshed to Fraser 2025 and (for Fraser-unavailable countries) TI CPI 2024 with the documented band convention. For pitch decks active as of v1.1 publication, the IC component should be regenerated from the v1.1 platform; the Scope statement makes the country-by-country composition transparent for jurisdictions where the IC scope previously implied broader country coverage than was sourced.
The transaction comparables (15 African M&A deals (within 25 comparables) 2015–2025) are unchanged in v1.1. The Score adjustments do not change deal comparable data. What does shift is the comparable-province Score against which a target’s province is benchmarked: provinces with materially downward v1.1 Scores recalibrate the benchmark deal-flow narrative for any active capital raise.
The v1.1 publication is the strongest demonstrable evidence to date of the platform’s correction discipline operating in public. Seven correction layers, attributed separately. 12 of 13 v1.0 Scores algebraically corrected. Three institutional source-citation cascades (Acquirer Brief, Category Paper, Independence Policy mirror) executed in lockstep. The discipline is not theoretical; it is operating in observable ways and producing visible outputs.
For acquirer-class evaluation: the v1.1 corrections are the kind of correction event that a competing platform built on undocumented methodology cannot survive — and that a platform built on documented methodology is structurally designed to absorb. The asset is the discipline, not the absence of corrections.
Several items that v1.1 explicitly does not touch:
Three items are committed to v1.2 development per §09 of the methodology document:
Four IAI values (DRC, Namibia, Angola, Egypt) carry PROVISIONAL status pending direct PDF retrieval from the Fraser Institute publication. The rank and bracket-derived estimates are source-cited; the exact decimals will land in a v1.1.1 patch with corresponding audit log entry on verification. v1.1.1 will publish Score values to one decimal where the exact-decimal verification changes the displayed Score; provinces where bracket-derived and exact-decimal IC values round identically will pass through without surface change.
The 30-day vintage-refresh SLA from each new Fraser publication is reaffirmed for future refreshes. The v1.1 exception window is closed; the next refresh (Fraser 2026, expected Q1 2027) will publish within 30 days of the Fraser release.
The v1.1 audit log entry is the canonical operational record — the source-of-truth document that names every correction layer, every input source, every cascading sub-entry, and every regression check. It is published at /audit-log and accessible to any reader who wants the full operational record.
This companion document exists because the audit log entry is structured for the operational reader (engineers, auditors, acquirer DD teams). A different reader — a DFI investment officer, an exploration geologist, an institutional reader evaluating the platform’s discipline — benefits from the same content rendered in different language: the corrections explained in plain terms, with verification paths surfaced, and the implications for each user role spelled out.
Both documents say the same thing. Neither is the more authoritative version; they are the same record at two reading depths.
Editorial responsibility for the v1.1 publication and all corrections rests with Nikesh Patel, Honorary Consul of Rwanda in Mauritius. Direct contact: nikesh@afrimintel.com.
Material errors flagged by any user are corrected within 7 business days per the Quality Standard. Non-material errors within 30 days. Data error reports, source attribution corrections, and methodology questions are all welcome via the same channel.
Afrimintel Africa Mineral Intelligence · Methodology v1.1 · 16 May 2026 Editorial responsibility: Nikesh Patel, Honorary Consul of Rwanda in Mauritius nikesh@afrimintel.com · afrimintel.com
This companion document is published alongside Methodology v1.1 and the v1.1 audit log entry. Discrepancies should be reported to nikesh@afrimintel.com; the audit log entry is the canonical operational record.