UAE Healthcare SEO Index Methodology

MAPS OF ARABIA / RESEARCH
UAE HEALTHCARE WEBSITE RESEARCH

UAE Healthcare SEO Index methodology

How the partially source-checked historical search ranking is calculated, which providers qualify, and how the evidence supports the results. All 50 provider profiles remain available.

45meet query coverage
137worksheet queries
10minimum supported queries
29 OCTOBER 2026 CORRECTIONPARTIALLY SOURCE-CHECKED HISTORICAL SEARCH

Research team

This research examines healthcare websites and search visibility. It does not assess clinical care.

01

What the index measures

The UAE Healthcare SEO Index retains all 50 provider profiles. Of these, 45 meet the minimum of 10 supported queries and 32 receive a displayed rank from a positive recorded search score. Thirteen coverage-eligible profiles have no recorded top-10 contribution and their rank is not established. This is a partially source-checked historical search view of exact domains, not every branch, related website or business within a healthcare group.

A displayed search rank requires at least 10 supported queries and a positive recorded search score. JTS, KIMS and Kanad retain their profiles with ranks withheld because coverage is below that threshold. Elaj and Private Care remain unranked because they have no supported-query denominator. Missing scores remain unavailable. Internal arithmetic zeros are displayed as “No recorded top-10 contribution”; a coverage-eligible zero has “Rank not established”. The index does not assess clinical quality, patient safety, traffic or conversion performance, and should not guide a patient’s choice of provider.

02

Historical four-component model

Historical composite model
ComponentWeightMeaning
Recorded search visibility60%The average position score across queries supported as relevant to the provider.
Laboratory speed20%Mobile and desktop performance, with a mobile loading-time penalty.
Website trust signals15%The percentage of eight website checks recorded as Yes.
One local-search snapshot5%Appearance in one retained three-listing Local Pack.

Historical composite = 0.60 × search + 0.20 × speed + 0.15 × trust signals + 0.05 × local snapshot. Component scores are on a 0–100 scale. A missing required input or an empty supported-query denominator yields no historical total. Weights are not redistributed. The 60/20/15/5 weights and non-search inputs are unchanged; this formula is retained for historical reconciliation. The current public view ranks search scores only and has no current composite rank.

03

Which queries enter a search score

The scoring worksheet contains 137 queries. Each provider’s denominator includes only queries supported as relevant to its services, setting and location. Queries with unresolved relevance remain separate from those documented as irrelevant. Supported, unresolved and irrelevant counts sum to 137 for each provider.

Within the originally selected worksheet run, accepted row-level corrections are applied before choosing the best supported position for each provider and supported query. The final exclusion step removes 31 unsupported positive contributions after the accepted position corrections, without changing any supported-query denominator. Position one earns 100 points, falling by ten points per position to ten points at position ten; later positions earn zero. The calculation is max(0, 110 − 10 × position), averaged equally across the fixed supported-query set. A query with no accepted recorded top-10 contribution adds no points to that arithmetic; unobserved results remain unknown, not verified absence.

The source corrections are applied to the selected rows. The 894 accepted source rows yield 447 selected supported rows and 446 positive provider–query contributions; the 31 unsupported positives are excluded. Records without verified absolute organic positions remain identified as unresolved evidence. Where the worksheet has several runs, the frozen selected run remains unchanged; source order does not establish capture chronology. No best-across-runs sensitivity or provisional page-two ordinal is promoted into the current score.

The current eligibility threshold is 10 supported queries. JTS and KIMS each have one supported query and 136 unresolved queries; Kanad has six supported queries and 131 unresolved. Their profiles remain available, but none receives a current search rank. Queries whose relevance remains unresolved stay outside the supported-query denominator and are not classified as irrelevant.

04

Historical speed component

The raw speed result is 60% of the mobile Lighthouse performance score plus 30% of the desktop score, minus a penalty for mobile Largest Contentful Paint (LCP). The penalty is zero at up to 2.5 seconds, ten points above 2.5 through four seconds, and 25 points above four seconds.

Speed = max(0, 0.60 × mobile performance + 0.30 × desktop performance − LCP penalty) ÷ 90 × 100. These are laboratory measurements, not a distribution of real-user performance. Original raw values are retained.

05

Historical trust checklist

The eight equally weighted checks are author bylines, a medical reviewer, a last-reviewed date, citations or references, an editorial-policy URL, a privacy-policy URL, Organization structured data and MedicalWebPage structured data. Every field requires Yes or No. The component score is the percentage marked Yes.

The checklist shows which features were found on a website. It does not verify medical accuracy or quality of care. Checking whether a website names a medical reviewer does not mean this Index has had clinical review.

06

Historical local snapshot and sensitivity

The historical local component uses one complete three-listing pack recorded on 15 April 2026 for “Best private Hospital”, with visible UAE attribution. International Modern Hospital, American Hospital Dubai Science Park and Medeor Dubai appear in that retained pack. Its exact origin, capture timestamp and activation of the visible filter remain unverified. A conflicting February snapshot is not combined with it.

In the historical composite, each domain matched to a listing in this complete pack receives 100 local-component points, contributing 5 points to that total. All other domains in the cohort receive zero for this pack. The three listings receive equal credit: position within the pack does not change the score. A missing required component is not scored as zero: it leaves the total uncalculated, with no redistribution of weights.

  • International Modern Hospital → imh.ae
  • American Hospital Dubai Science Park → ahdubai.com
  • Medeor Hospital Dubai → medeor.ae; no additional credit goes to the parent domain, burjeel.com.

This observation does not measure nationwide or current local visibility. In the original historical composite, removing only the five-point local contribution, without redistributing or rescaling the other weights, moves Medcare ahead of American Hospital Dubai. The hypothetical maximum becomes 95 points. This historical sensitivity example does not change the 60/20/15/5 weights or determine the current search-only ranking.

07

Dates and component selection

The search records date from April–May 2026. Complete original speed and trust records take precedence over later supplements. Historical measurement dates remain unknown where the records do not provide them.

October supplements fill specific gaps: speed for Burjeel and SEHA from the 19–21 October collection window, and for Dubai Health from 19 October; trust for Aster, Burjeel, Dubai Health and SEHA from the 19 October audit. Most service-scope evidence was retrieved on 19 October. The JTS urgent-care assessment uses cached official text retrieved on 20 October. These scope checks are not backdated to the historical searches.

Al Sharq has two historical speed records. The study uses the first in source order; it is not known to be newer or more accurate. In the original historical composite, the retained alternative adds 0.6 points and changes its position by three places. Both records remain documented; this comparison does not change the current search-only ranking.

08

Current search ranks and separate language panels

Exact, unrounded positive recorded search scores determine competition ranks among 32 displayed providers. Forty-five profiles meet query coverage, but the 13 eligible profiles with no recorded top-10 contribution have no established rank. Equal exact positive scores share a rank and subsequent ranks skip accordingly. Domain order is used only for stable display, never to break a tie. Positive scores display to two decimal places using half-up rounding; rounding does not determine rank. Suppressed, withheld or unranked profiles never fall back to a numerical-only or historical composite rank.

The separate reconstructed panels contain 54 English sequences and 540 observations, and 47 Arabic sequences and 470 observations. A domain counts once per query in reconstructed reach. These counts enter neither the current search ranking nor the historical composite, and do not replace the supported-query denominator. Different query sets prevent a combined bilingual league table or a language-performance conclusion from a simple percentage difference.

Panel selection and evidence coverage

Panel selection and scope

This edition preserves 137 distinct query strings from the team's historical worksheet and 50 provider-domain profiles. The full query list contains 75 Arabic-script queries and 62 queries without Arabic script. It includes general hospital discovery, specialties, appointments and place-modified searches. Near-equivalent phrases remain separate entries, as they were in the retained worksheet. They are not weighted by search volume.

Each profile refers to its listed domain. Related hospital, group and subsidiary domains are not automatically merged. The panel records the providers and queries available in the retained research materials; the recovered sources do not establish selection probabilities, comprehensive market coverage or a complete original inclusion/exclusion rationale. Results should therefore be read within this panel, without assuming that it represents all healthcare searches or providers in the UAE.

Read coverage with the score

The 10-query eligibility threshold counts queries with positively supported relevance to the provider. It does not require 10 observed or independently verified search positions.

For each provider, the coverage table separates: queries with an admitted positive contribution; queries recorded in the worksheet but without admitted proof for the selected position; and supported queries with no worksheet appearance. These three counts add to the provider's supported-query denominator. The last two states contribute no points to the recorded-evidence score. Neither proves absence from search.

American Hospital's denominator of 93 comprises 55 admitted positive-query contributions, 30 recorded queries without admitted proof and 8 with no worksheet appearance. Zulekha's denominator of 99 comprises 12, 57 and 30 respectively. Different evidence coverage and different relevant-query sets limit direct comparison between their recorded scores.

What source admission establishes

The score uses admitted evidence matched to selected worksheet records. It is not a reconstruction of every provider visible in every retained screenshot. A provider can be visible in an image yet receive no contribution when the selected worksheet record is absent or points to a different source. For example, the retained CAN137-001 image shows Cleveland, Burjeel and Aster as well as the admitted King's, Emirates and American observations; the current ledger does not admit those additional three appearances. The published score and rank remain unchanged.

This distinction matters when interpreting the table: it orders the contributions admitted to this historical ledger. It does not establish complete organic-search coverage, current market leadership or traffic share. A broader reconstruction would require a documented admission rule applied consistently across the whole panel, a versioned recalculation and a new release review.

Source image for the CAN137-001 example: https://drive.google.com/file/d/1rxwBG4hK2uHRF8UdEY-mn4q7Slfj-Ntd/view

Complete query panel: 137 queries

Query IDs follow the first appearance of each exact phrase in the native worksheet. Dates below are worksheet labels, not authenticated search-session timestamps. The existing calculation dictionary contains 134 queries; CAN137-025, CAN137-076 and CAN137-077 are included here to complete the panel and do not change the 3,417 published supported-pair calculations.

Show all 137 queries
Complete query panel: 137 queries
Query IDExact source queryFirst worksheet rowWorksheet date
CAN137-001top hospitals in uae42026-04-13
CAN137-002hospitals in ajman uae402026-04-13
CAN137-003best cancer hospital in uae922026-04-13
CAN137-004best eye hospital in uae1412026-04-13
CAN137-005best neurology hospital in uae2162026-04-13
CAN137-006best oncology hospital in uae2432026-04-13
CAN137-007best urology hospital in uae2692026-04-13
CAN137-008ent hospitals in uae2982026-04-13
CAN137-009famous hospitals in uae3342026-04-13
CAN137-010best cardiac hospital in uae3872026-04-13
CAN137-011best hospital for ivf in uae4062026-04-14
CAN137-012best orthopedic hospital in uae4512026-04-14
CAN137-013gastric hospital in uae5282026-04-14
CAN137-014hospital in the uae5732026-04-14
CAN137-015hospital in uae dubai6032026-04-14
CAN137-016list of all hospitals in uae6522026-04-14
CAN137-017list of private hospitals in uae6722026-04-14
CAN137-018neurology hospitals in uae6952026-04-14
CAN137-019best cardiology hospital in uae7242026-04-14
CAN137-020cardiac surgery hospitals in uae7492026-04-14
CAN137-021heart hospital in uae7742026-04-15
CAN137-022Private hospital8262026-04-15
CAN137-023Best private Hospital9242026-04-15
CAN137-024Best hospital in UAE9622026-04-15
CAN137-025private hospital around me10562026-04-15
CAN137-026nephrology hospitals in uae11012026-04-15
CAN137-027private hospitals in uae12012026-04-15
CAN137-028psychiatric hospitals in uae12302026-04-15
CAN137-029best eye hospitals in uae12662026-04-15
CAN137-030best hospitals in dubai uae12952026-04-15
CAN137-031مستشفى خاص13402026-04-21
CAN137-032أفضل مستشفى خاص14342026-04-21
CAN137-033أحسن مستشفى خاص15342026-04-21
CAN137-034مستشفى16342026-04-21
CAN137-035أفضل مستشفى17282026-04-21
CAN137-036أحسن مستشفى18272026-04-21
CAN137-037حجز موعد مستشفى19262026-04-21
CAN137-038حجز موعد في مستشفى20262026-04-21
CAN137-039استشارة مستشفى21262026-04-21
CAN137-040حجز استشارة مستشفى22262026-04-21
CAN137-041تقييمات مستشفى22802026-04-21
CAN137-042تقييمات مستشفى خاص23302026-04-21
CAN137-043مستشفى يقبل التأمين24302026-04-21
CAN137-044تأمين مستشفى25102026-04-21
CAN137-045مستشفى تأمين صحي26072026-04-21
CAN137-046رقم مستشفى27042026-04-21
CAN137-047رقم هاتف مستشفى28022026-04-21
CAN137-048عنوان مستشفى29012026-04-21
CAN137-049مواعيد زيارة المستشفى30002026-04-21
CAN137-050حجز اونلاين مستشفى30702026-04-21
CAN137-051مستشفى طوارئ31702026-04-21
CAN137-052قسم الطوارئ32642026-04-21
CAN137-053طوارئ ٢٤ ساعة33582026-04-21
CAN137-054مستشفى ٢٤ ساعة طوارئ34572026-04-21
CAN137-055رعاية عاجلة35542026-04-21
CAN137-056طبيب طوارئ مستشفى36532026-04-21
CAN137-057إسعاف مستشفى37532026-04-21
CAN137-058مستشفى إسعاف38502026-04-22
CAN137-059مستشفى ولادة39432026-04-22
CAN137-060أفضل مستشفى ولادة40372026-04-22
CAN137-061نسائية وتوليد مستشفى41362026-04-22
CAN137-062طبيبة نسائية في مستشفى42362026-04-23
CAN137-063مستشفى أطفال43362026-04-23
CAN137-064طبيب أطفال في مستشفى44312026-04-23
CAN137-065مستشفى قلب45312026-04-23
CAN137-066طبيب قلب في مستشفى46272026-04-23
CAN137-067مستشفى عظام47272026-04-23
CAN137-068طبيب عظام في مستشفى48212026-04-23
CAN137-069مستشفى جراحة49202026-04-23
CAN137-070جراحة بالمنظار مستشفى50172026-04-23
CAN137-071أنف وأذن وحنجرة مستشفى51142026-04-23
CAN137-072فحص سمع مستشفى51842026-04-23
CAN137-073مستشفى جلدية52842026-04-23
CAN137-074مستشفى عيون53802026-04-23
CAN137-075hospitals in ras al khaimah uae54752026-04-21
CAN137-076hospitals in umm al quwain uae54882026-04-21
CAN137-077largest hospital in uae55852026-04-21
CAN137-078leading hospitals in uae56822026-04-21
CAN137-079major hospitals in uae57762026-04-21
CAN137-080best hospital in dubai uae58742026-04-21
CAN137-081best heart hospital in uae59732026-04-21
CAN137-082ليزك مستشفى60502026-04-23
CAN137-083مستشفى مسالك بولية61492026-04-23
CAN137-084غسيل كلى مستشفى62492026-04-24
CAN137-085ألم معدة مستشفى63462026-04-24
CAN137-086مستشفى سرطان64452026-04-24
CAN137-087رنين مغناطيسي مستشفى65412026-04-24
CAN137-088مستشفى تأهيل66122026-05-03
CAN137-089مختبر تحاليل مستشفى67112026-05-03
CAN137-090سونار مستشفى69112026-05-03
CAN137-091سرطان الثدي مستشفى70482026-05-03
CAN137-092علاج طبيعي مستشفى71482026-05-03
CAN137-093جراحة عامة مستشفى72472026-05-03
CAN137-094مستشفى حديثي الولادة73472026-05-03
CAN137-095ولادة طبيعية مستشفى74452026-05-03
CAN137-096ولادة قيصرية مستشفى75452026-05-03
CAN137-097كبد مستشفى76332026-05-03
CAN137-098ألم ظهر مستشفى77322026-05-05
CAN137-099فحص قلب مستشفى78292026-05-05
CAN137-100عملية فتق مستشفى79272026-05-05
CAN137-101عملية الزائدة مستشفى80262026-05-05
CAN137-102علاج إكزيما مستشفى81212026-05-05
CAN137-103حصى الكلى مستشفى82192026-05-05
CAN137-104بروستات مستشفى83162026-05-05
CAN137-105مستشفى أورام84132026-05-05
CAN137-106قسم الأشعة مستشفى85122026-05-05
CAN137-107تحاليل دم مستشفى85922026-05-05
CAN137-108تأهيل بعد الجراحة مستشفى86922026-05-05
CAN137-109إعادة تأهيل مستشفى87892026-05-05
CAN137-110منظار معدة مستشفى88892026-05-05
CAN137-111علاج كيماوي مستشفى89872026-05-05
CAN137-112فحوصات مخبرية مستشفى90862026-05-05
CAN137-113top private hospital91852026-05-12
CAN137-114best hospital92562026-05-12
CAN137-115top hospital93552026-05-12
CAN137-116private hospital appointment94532026-05-12
CAN137-117hospital consultation95472026-05-12
CAN137-118hospital insurance96452026-05-12
CAN137-119private hospital insurance97442026-05-12
CAN137-12024 hour emergency hospital98442026-05-12
CAN137-121urgent care hospital99382026-05-12
CAN137-122maternity hospital100332026-05-12
CAN137-123gynecology hospital101282026-05-12
CAN137-124pregnancy care hospital102252026-05-12
CAN137-125pediatric hospital103132026-05-12
CAN137-126orthopedic hospital104082026-05-12
CAN137-127back pain hospital105062026-05-12
CAN137-128general surgery hospital106022026-05-12
CAN137-129surgery hospital106992026-05-12
CAN137-130dermatology hospital107952026-05-12
CAN137-131urology hospital108902026-05-12
CAN137-132kidney stone hospital109442026-05-12
CAN137-133gastroenterology hospital110402026-05-12
CAN137-134cancer hospital111382026-05-12
CAN137-135radiology hospital112332026-05-12
CAN137-136physiotherapy hospital112792026-05-12
CAN137-137post surgery rehab hospital113782026-05-12

Native source worksheet

Appendix published 11 November 2026. Research observation dates, scores and rankings are unchanged.

09

Evidence access and interpretation

The approved correction dataset retains all 50 provider profiles, exact internal search arithmetic, public display states, historical component values and provenance coverage. Its SHA-256 is d2bba986d8cd8971984e3923bdb44e208a3e17012337b0b115b5922b5c963da3. The hash identifies this calculation dataset; it does not certify capture conditions. Historical composite scores are retained separately and are not relabelled as corrected public composite scores.

Full replication needs the underlying records and calculation rules, not only the report or its hash. Some source files remain private. Requests for a specific record remain subject to existing access and redistribution permissions. Mixed periods, unequal query coverage and recorded-position conventions limit direct comparisons.

From the original worksheet to the current search score

Two operations separate the historical search column from the current score: accepted position corrections, then exclusion of unsupported positive contributions. The archived 27 October 2026 edition records the intermediate calculation. Its methodology states that accepted row-position corrections were applied while unadmitted rows retained their historical positions (pages 8–9). Its frozen ledger identifies 186 original-source observations, 708 alternate-source observations and 413 changed provider/query contributions (page 42). This is version history for the same historical dataset, not a new measurement.

Reconstruction from the frozen row register gives the original worksheet 422 positive provider/query contributions. Position corrections change 413 provider/query point values and produce 477 positive contributions in the intermediate calculation documented by the 27 October edition. The final exclusion removes 31 unsupported positives, leaving 446. Each provider keeps the same supported-query denominator.

American Hospital Dubai illustrates the bridge: 2720/93 ≈ 29.25 originally, 4280/93 ≈ 46.02 after position corrections, and 4100/93 ≈ 44.09 after exclusions. These are three calculations over the same historical query set, not growth over time.

Tadawi’s original 40/84 came from two queries: native row 533, CAN137-013, contributed 10 points at worksheet position 10 and is now excluded; native row 11044, CAN137-133, contributed 30 points at worksheet position 8 and now contributes 60 at admitted ordinal 5. The bridge is therefore 40/84 → 70/84 → 60/84, not a single position-7-to-5 change.

All 32 currently ranked providers differ between the original and final calculations. The stated 17 changed scores, six changed numerical ranks and unchanged top-ten order compare only the 27 October position-corrected calculation with the final exclusion step. They do not establish stability from the original worksheet, across other source choices, or in current search results.

Position corrections are row-specific; no uniform offset is applied. A match between worksheet and source rows does not authenticate the capture time, session or all absolute organic positions. Existing source and coverage limitations remain in force.

Correction scope and remaining evidence gaps

Compared with the position-corrected calculation published in the 27 October 2026 edition, the final exclusion step removes 31 unsupported positive provider–query contributions. This final step changes 17 internal search scores and six numerical-only rank values; separately, 13 former rank-33 labels are suppressed, so 19 public rank fields differ. Across this final exclusion step, the top-10 order is unchanged and no provider becomes newly zero. All 50 profiles, supported-query denominators, historical composites, speed, trust, local inputs and collection dates remain unchanged. The 894 accepted source rows comprise 186 original-source and 708 accepted alternate-source observations. Alternate captures do not authenticate the original screenshot identity or session. There are still 974 selected records without verified absolute positions. Of these, 408 have page-local content matches, which do not establish absolute organic positions. Unobserved results are unknown, not verified absence. Queries whose applicability remains unresolved stay outside the denominator. This remains a partially source-checked historical view, not a fully verified organic ranking.

Evidence for this edition

Release: 29 October 2026. Historical collection dates unchanged.

Recorded search score excludes 31 unsupported positive contributions. The 894-row source ledger yields 446 selected supported positive query contributions.

Inspect queries and sources

Calculated over the fixed historical query set. Unobserved results are unknown, not verified absence.

974 selected records still lack verified absolute positions. Of these, 408 have page-local content matches, but their absolute positions remain unverified. No recorded appearance is not proof of absence. Alternate captures do not authenticate the original session.

The targeted archive review covered 262 of 737 files. Page-local matches, uncertain paid-result classifications and unmatched records are excluded from score corrections. Page-local traceability reviewed 27 October 2026; no page-local positions were added to scoring.

Source files retain their existing access settings. On 27 October 2026, all 125 linked source viewers and their PNG files were accessible without signing in; the downloaded files matched the retained source hashes.

Calculation dataset SHA-256d2bba986d8cd8971984e3923bdb44e208a3e17012337b0b115b5922b5c963da3

Read the complete comparison

Explore the 2026 findings · Download the full report

For factual corrections, email info@mapsofarabia.com with the domain, passage or result, and supporting source. Do not send patient information or other unnecessary sensitive data.

Maps of Arabia.
Back to top