The Harch 100 is our ranking of Morocco's 100 most reputable companies. It is not a popularity contest, a media-mention count, or an editor's pick. It is a scored index built from four data inputs, three pillars, and a human expert review. This is the methodology, and a worked example for Bank of Africa at 72/100.
The scoring formula
The headline score is a weighted composite of four inputs: media sentiment (40%), AI-engine citation sentiment (25%), social-platform conversation tone (20%), and expert-panel review (15%). Each input is normalised to a 0–100 scale and combined linearly. The formula is intentionally simple — the complexity lives in the input measurement, not the combination.
Pillar extraction
Behind the headline score, we decompose reputation into three pillars: Innovation (the future-facing narrative), Performance (delivered results), and Purpose (social and environmental frame). Pillar scores are extracted from the same input data but with different query and weighting logic — Innovation weights forward-looking language and R&D signals, Performance weights financial and operational coverage, Purpose weights ESG and community-impact coverage. The pillar mix is often more diagnostic than the headline.
Data sources
- 30+ Moroccan and pan-African media sources, scraped via Google News RSS and direct publisher feeds, processing 5M+ articles per day.
- Eight AI engines — ChatGPT, Perplexity, Gemini, Claude, Copilot, Mistral, Google AI Overviews, Grok — queried with a 240-prompt battery quarterly.
- Social platforms (X, Facebook, Instagram, YouTube, TikTok) monitored via platform APIs and partners, with Darija-native sentiment.
- An expert panel of nine Moroccan senior comms, IR and risk professionals who review the top 20 and bottom 10 each cycle.
Worked example: Bank of Africa at 72
Bank of Africa's 2026 Q1 score is 72. Here is how it is built. Media sentiment, measured across 842 articles in the quarter, normalises to 70. AI-engine citation sentiment, across 60 prompts and four engines, normalises to 68. Social-platform tone, across 12,400 mentions with Darija-native classification, normalises to 74. Expert-panel review, on a 0–100 rubric, scores 78.
Pillar decomposition for Bank of Africa
Bank of Africa's pillar mix tells a sharper story than the headline 72. Innovation scores 78 (strong sustainability and digital-banking narrative), Performance scores 66 (restructuring noise weighs), Purpose scores 80 (sustainable-finance framework leads the sector). The diagnosis: the score is capped not by narrative ambition but by performance execution. The levers that move the score are operational, not communicational.
What we exclude
The Harch 100 excludes paid placement, sponsored content, and any coverage we can identify as commissioned. We exclude employee-review platforms from the headline score (they are tracked separately, as a workplace sub-index). We exclude AI-generated content farms. The objective is a measure of earned reputation — what the world says about a company, not what the company pays to have said.
The expert review
The 15% expert weight is deliberate. Quantitative inputs are necessary but not sufficient — they miss context, nuance, and the difference between volume and significance. Our panel of nine senior practitioners reviews the top 20 and bottom 10 each cycle, scoring on a 0–100 rubric that accounts for crisis handling, narrative coherence, and stakeholder trust signals the automated inputs cannot capture. The panel's score is the corrective against gaming.
How to read the ranking
The headline score is a snapshot. The pillar mix is the diagnosis. The quarter-on-quarter delta is the trajectory. A company at 72 with a rising Innovation pillar and a stable Performance pillar is on a different trajectory from one at 72 with both pillars flat. The Harch 100 is most useful when read as a movement, not a podium.
The full methodology, including the source list, the prompt battery, the normalisation logic and the expert-panel rubric, is published in our method documentation. Harch Atelier clients receive the sector decomposition and the company-level scorecard underlying their ranking.