1.67 million pilgrims. 171 countries. 5 days. Hajj is the world's largest annual gathering — and one of the most complex urban management challenges on the planet. Traditional metrics count crowds and incidents. But how do you measure whether the experience itself is actually improving?
The Challenge: Experience at an Unprecedented Scale
Managing Hajj logistics is an extraordinary achievement of urban planning and crowd management. But logistical success and experiential quality are not the same thing. A pilgrim may arrive safely, be accommodated adequately, and complete the rituals without incident — and still feel that the experience fell short of expectations.
Traditional approaches to measuring satisfaction at this scale are impractical. Face-to-face surveys cannot reach a meaningful sample across 171 nationalities in five days. Post-event questionnaires suffer from recall bias and low response rates. And anecdotal feedback, while valuable, cannot be systematically compared across years or locations.
The question we set out to answer: could AI-driven sentiment analysis provide a continuous, multi-language, spatially-anchored measure of pilgrim satisfaction — and could it do so at a granularity that enables targeted intervention?
The Methodology Applied to Hajj
Four capabilities were brought together for this analysis:
Arabic-native sentiment at the core
The AI models were trained specifically for Arabic expression, including the dialects spoken across the Gulf. But Hajj is multilingual by nature — pilgrims post in over 100 languages. The system processes Arabic, English, Urdu, Indonesian, Turkish, Malay, and dozens more, applying sentiment scoring calibrated for each linguistic context.
Location-based mapping
Every expression of satisfaction or frustration was anchored to a precise location within the Hajj journey — from pre-departure preparation through arrival, Ihram, Tawaf, Sa'i, Arafat, Muzdalifah, Jamarat, and departure. This spatial anchoring transforms general feedback into location-specific intelligence.
Topic-specific analysis
Rather than producing a single satisfaction number, the analysis breaks down into the specific issues that drive dissatisfaction: transport, accommodation, crowd management, wayfinding, food quality, healthcare access, safety, and spiritual guidance. Each topic receives its own score across each stage of the journey.
Temporal comparison
The same methodology was applied to both Hajj 2024 and Hajj 2022, enabling a direct year-on-year comparison of performance across every topic and location.
What the Analysis Revealed
The 2024 vs 2022 comparison surfaced clear patterns of both improvement and persistent challenge.
Measurable improvements
Accommodation satisfaction improved by 12 percentage points. Food quality rose by 8 points. These gains suggest that targeted investment between the two Hajj seasons had a measurable impact on pilgrim experience.
Persistent challenges
Transport satisfaction remained 5 points below threshold. Crowd management sat 11 points below threshold. These are the areas where further intervention is needed — and the data now provides the specificity to act on them.
The experience gap between demographics
By segmenting the analysis between Saudi and non-Saudi pilgrims, the data revealed a consistent experience gap across every service category. Non-Saudi pilgrims reported lower satisfaction in housing, transport, public spaces, technology, and social and community aspects. Saudi pilgrims scored higher in most areas except healthcare services, where non-Saudi pilgrims reported greater satisfaction — and spiritual guidance, where non-Saudi pilgrims also scored higher.
This demographic segmentation enables service providers to identify where the experience diverges and design interventions that address specific population needs rather than applying blanket improvements.
From Data to Targeted Interventions
The power of spatially-anchored, topic-specific satisfaction data is that it enables precise action rather than general improvement efforts. Instead of a mandate to “improve everything,” the analysis produces recommendations such as:
- Location: Bus interchange Zone 3
- Topic: Crowd management and wayfinding signage
- Action: Additional marshals, multilingual signage, queue management
This shifts the conversation from “improve transport” to “fix this specific issue at this specific place.” Resources can be allocated where they will have the greatest impact on the actual pilgrim experience.
Why This Matters Beyond Hajj
The Hajj case study is an extreme test of this methodology — massive scale, compressed timeframe, multilingual participants, extreme environmental conditions. If the approach can produce actionable insights here, it can be applied to any urban context where understanding lived experience matters.
The same framework has been applied to city-scale analysis across the GCC, measuring resident satisfaction with housing, transport, public spaces, and community services on an ongoing basis. The principles are identical: transform unstructured human expression into structured, spatial, actionable intelligence.
“Sentiment without verification is opinion. Verified sentiment is evidence for action.”
This article is part of a series based on the “Reading the City” presentation at the 61st ISOCARP World Planning Congress. The companion articles cover the full methodology framework and a second case study examining how AI-driven analysis shaped housing guidelines in Al Ain.
See Sentiment Analysis in Action
The satisfaction scoring methodology described in this case study is available through the GUS platform. Explore how AI-driven urban intelligence can inform your planning decisions.