Skip to main content
RoastIQBuyerLensHugoPricingBlogAbout
Book a demoSign inStart free →
Market insight · n=296 · published 18 May 2026 · 10 min read

The MENA creative boom.

Arabic-first ads score +14 composite over EN-localised across our six-market MENA cohort. Darija TikTok lifts the Moroccan cohort +1.7σ. MSA YouTube lifts Saudi +0.9σ. The region is not "Western markets translated", it has its own scroll-stop, its own ritual hooks, and a creator economy maturing fast.

Sample · n=296 Markets · 6 Language strategies · 3 Median scroll-stop · 2.4s Confidence · MED
Oussama Nakhil portrait
Oussama Nakhil · Founder & CEO
Multiple years buyer-side: NielsenIQ insights, then L'Oréal Groupe in global marketing insights. Moroccan founder, Casablanca-based. Every chart pinned to model_version and benchmark_pool_version v.2026-05.
+14
Arabic-first composite lift vs EN-localised
+1.7σ
Darija TikTok · Morocco cohort
2.4s
Median MENA scroll-stop · vs 1.9s global

01 · The headline: Arabic-first wins.

Across our 296-ad MENA cohort spanning Morocco, Algeria, Tunisia, Egypt, Saudi Arabia and the UAE, ads built Arabic-first, written in the appropriate dialect register, shot with native talent, edited to local cadence, score +14 composite points over EN-localised cuts. The lift is concentrated in Beat the Skip (+11) and Build Brand (+18), which together drive the composite. Sell Proposition and Brand Impact also move, more modestly.

"EN-localised" in our taxonomy means a cut originally produced for an English-speaking market and translated into Arabic (subtitle, dub or copy swap) without re-shoot. It is the failure mode of the multinational that treats MENA as an extension of its EMEA plan rather than as its own creative market. The format is widespread, easy to detect, and reliably the lowest-scoring of the three strategies in our cohort.

Arabic-first composite
68
Median · n=176 across 6 markets
MSA composite
62
Median · pan-Arab register
EN-localised composite
54
Median · translated-only cuts
MENA scroll-stop
2.4s
Median · vs 1.9s global
The brief

"Localised copy is not localised creative. Translation gives you the words; dialect, ritual and cadence give you the lift. The +14 is the gap between the two."

02 · The language strategy table.

The strategic choice is rarely "Arabic or English"; it is "which Arabic." Darija, MSA and Khaleeji are not interchangeable registers, and the lift varies sharply by (market × surface × register). The table below uses the matched-cohort method (same brand, same product, multiple register variants where available) to isolate the register effect from the brand and category effects.

// Composite by language strategy × surface (n=296)

Local dialect is the strongest signal in our cohort. EN-localised is the floor across every surface measured.
Local dialect · TikTok
74
MSA · TikTok
62
EN-localised · TikTok
48
Local dialect · Meta-Feed
68
MSA · YouTube
66
EN-localised · YouTube
55

Two patterns matter. First, the local-dialect lift is largest on TikTok (the surface where native creator cadence is the most discriminating signal) and smaller on YouTube (where MSA's pan-Arab reach offsets the dialect specificity). Second, EN-localised never tops 55 on any surface in our cohort. It is the structural floor, the cut "works" only by the loose definition of "does not actively harm the brand," which is a low bar for a paid distribution dollar.

03 · The six markets, profiled.

MENA is not one market. The strategic register, the scroll-stop median, the dominant local brands and the platform mix vary by country. Each market profile below summarises the working answer for that market based on our 2026-Q1/Q2 cohort.

// MAMorocco, Darija dominant
Sample · 62Scroll-stop · 2.4sDominant register · Darija

Morocco is the Darija stronghold of our cohort. Darija TikTok carries a +1.7σ composite lift over MSA and a +2.6σ lift over EN-localised. The dialect is informal, code-switches freely with French and standard Arabic, and rewards conversational creator-direct delivery. Inwi, IAM and Marjane lead the local-brand pool; the strongest cuts treat Casablanca and Marrakech street cadence as the creative ground truth, not a stylistic flourish.

French-only cuts underperform Darija by a margin similar to EN-localised but with a smaller absolute floor, the legacy of the bilingual market creates a tolerance for French that the Saudi cohort does not extend to English. Mid-tier brands that ship French-and-Arabic dual cuts often discover that the Darija version outperforms by 12-18 points even when the French version is well-crafted.

// DZAlgeria, Darija + French sub-strand
Sample · 38Scroll-stop · 2.3sDominant register · Darija

The Algerian sub-cohort sits closely with Morocco on register dynamics, Darija leads, French co-exists, MSA reads as formal-broadcast. Median scroll-stop is 2.3 seconds; the cohort is smaller (n=38) so the magnitudes carry MEDIUM confidence, but the direction is stable. Telecoms (Djezzy, Ooredoo Algérie) and food brands lead the local-brand pool. Code-switched Darija/French cuts perform on par with pure Darija on TikTok; pure French cuts pay a measurable penalty.

// TNTunisia, Darija + French code-switch native
Sample · 34Scroll-stop · 2.2sDominant register · Darija / French

Tunisia is the most thoroughly code-switched market in our cohort. The Tunisian Darija is mutually intelligible with Moroccan and Algerian Darija but distinct in lexicon; French code-switching is so deeply native that pure Arabic cuts can read as "imported from elsewhere." Ooredoo Tunisia and local FMCG brands lead. The strongest cuts pace the code-switch deliberately, not stylistically, the switch reads as authentic register, not as a creative trope.

// EGEgypt, Egyptian Arabic, pan-MENA reach
Sample · 52Scroll-stop · 2.5sDominant register · Egyptian Arabic

Egyptian Arabic is the most regionally-influential dialect in MENA, Egyptian media (film, music, social) has built dialect familiarity across the region. Cuts in Egyptian Arabic carry secondary reach into Khaleeji and Maghrebi markets that pure-local-dialect cuts do not. The Egyptian sub-cohort scores +12 composite over EN-localised, with the cohort's slowest scroll-stop (2.5s) reflecting a longer-form storytelling tradition that biases the surface toward setup. Vodafone Egypt, Etisalat Misr and local FMCG dominate the local-brand pool.

// KSASaudi Arabia, Khaleeji, premium ad market
Sample · 60Scroll-stop · 2.2sDominant register · Khaleeji / MSA mix

Saudi is the highest-CPM market in MENA and the most strategic for global brands. Khaleeji dialect leads on TikTok and Snap; MSA carries more weight on YouTube and longer-format brand-build placements. STC, Almarai and Aramco dominate the local-brand pool. The Vision 2030 cultural shift is visible in the cohort, emotional and identity-led storytelling carries higher Build Brand scores than in any other MENA market we measure. EN-localised cuts pay the largest penalty in the cohort; the market reads English-only as a tell of low brand-priority.

// UAEUAE, Khaleeji / MSA / EN mix, expat-heavy
Sample · 50Scroll-stop · 2.2sDominant register · Khaleeji / MSA / English

The UAE is the only MENA market in our cohort where English-language cuts do not pay a structural penalty, the expat-heavy audience and the regional-hub positioning make English a legitimate native register for many brands. That said, Khaleeji and MSA cuts still carry a +8 composite lift on average. Etisalat, du and Emirates lead the local-brand pool. Multi-register campaigns (Khaleeji + English) outperform single-register on this market by 4-6 points across the cohort.

04 · Why MENA scrolls slower.

The MENA median scroll-stop is 2.4 seconds, half a second slower than the 1.9s global median. Three reasons converge to produce the gap, and the gap is stable across pool versions.

Cultural-ritual hook archetypes. The region's strongest creative tradition leans on ritual openings, the iftar table, the family gathering, the souk opening, the call-to-prayer sonic. These openings are slower by design and the viewer's attention is conditioned to honour the opening before the proposition lands. A 0.9-second hook reads as Western-style impatience on these surfaces; a 2.2-second hook reads as appropriate cadence.

Lower ad saturation per session. Despite high TikTok and Meta penetration, the median MENA viewer sees fewer ads per session than the median US viewer, particularly on premium placements. Lower saturation means lower pattern-recognition fatigue, which extends the median tolerance for setup. The effect is largest in Saudi and smallest in the UAE (which sits between MENA and EMEA on this measure).

Longer-form storytelling tradition. Egyptian cinema, Khaleeji drama, Moroccan music-video culture, the region's narrative arts bias the audience toward setup, character, and arc. The short-form vertical surfaces inherit some of that expectation. The viewer who scrolls through a 2.5-second opening on Egyptian TikTok is not being patient; they are being culturally consistent with the medium they grew up with.

05 · The MENA creator economy is maturing.

The creator-economy infrastructure in MENA crossed a structural threshold in 2025. Three trends are now visible in our cohort. First, Anghami, the regional music and audio platform, is now a creative-distribution surface in addition to a music service, and its creator-led commerce experiments are surfacing in branded-content briefs across Saudi and UAE. Second, MrBeast Arabic and the wave of regional dubbed creator content has built MENA-specific creator audiences at scale, with cohort-relevant brands beginning to invest in dubbed-creator partnerships. Third, local creator agencies (Boutiqaat in Kuwait, the Riyadh-based agency networks, Egypt's emerging influencer-marketing infrastructure) are absorbing the cadence and craft tradition that creator agencies in the US developed over 2018-2022, five years compressed into roughly eighteen months.

The implication for brands is that the supply side of the creator economy is no longer the bottleneck. The bottleneck is the brand's willingness to brief in dialect, hire local talent, and treat MENA as a creative market rather than a translated extension. Global brands that flip this assumption in the next 12-18 months will capture the +14 lift before the lift narrows.

06 · Four worked examples.

EX-01 Inwi · Darija street-stunt SCALE · 76
Market · MARegister · DarijaHook · Casablanca street stuntSurface · TikTok

A creator-led street-stunt cut shot in central Casablanca, native Darija delivery, ambient market sound, brand handshake via Inwi's distinctive purple at 0.8s. Beat the Skip 79, Build Brand 74. The cut hits all four hook moves and lands cultural-ritual authenticity from frame one. Among the highest Build Brand scores in the Moroccan sub-cohort.

EX-02 STC · Khaleeji emotional SCALE · 73
Market · KSARegister · KhaleejiSurface · YouTube + TikTokCue · 1.4s

A 30-second emotional cut for STC's connectivity proposition, native Khaleeji voice-over, family-table opening, distinctive STC purple at 1.4s. Build Brand 78, Brand Impact 74. The Vision 2030 emotional-identity tradition is visible in the cut's tonal craft. The EN-localised version of the same script scored 61 on the matched cohort, a 12-point gap on identical structural beats.

EX-03 MrBeast Arabic dub, branded SCALE · 74
Market · Pan-MENARegister · MSASurface · YouTubeFormat · Creator-led branded integration

A branded integration inside a MrBeast Arabic-dub episode, MSA voice-over to maximise pan-Arab reach, integration arrives at 12 seconds into the longer-form cut. Beat the Skip 78, Sell Proposition 71. The episode's distribution mechanic carries the cut past the scroll-stop discipline of TikTok; the brand's reward is the audience-trust transfer from the creator to the brand. The format is increasingly common in 2026 MENA branded content.

EX-04 Global FMCG · EN-localised failure REBUILD · 49
Market · MA / DZRegister · EN-localised (dubbed)Surface · TikTok + Meta-FeedHook · Western-pace

A global FMCG cut produced for the US market and dubbed into MSA for MENA distribution. Western 0.9-second hook, no cultural-ritual opening, brand handshake via wordmark at 4.0s. Beat the Skip 45, Build Brand 42. The cut works in the US; in Morocco and Algeria it reads as imported, fast, brand-late. The matched Darija-first version of a comparable brief in the cohort scored 67. The +18 spread is the cost of treating MENA as a translation market.

07 · Implications by stakeholder.

// For the CMO

If MENA appears as a line in your EMEA plan, you are leaving the +14 lift on the table by structure. Build a MENA brief, a MENA budget and a MENA cohort target. Hire dialect-native talent. The Saudi and UAE markets in particular reward this discipline at premium CPMs.

// For the brand team

Develop dialect-specific distinctive-asset guidelines, not just colour and wordmark rules. Inwi's Casablanca street cadence is a brand asset; STC's family-table emotional grammar is a brand asset. Treat the cultural-ritual hook as part of the brand book, not as a creative variable.

// For the agency

Stop briefing "Arabic dub" of US-produced cuts. Brief Arabic-first from the strategic brief through the shoot. Hire MENA creators on the same terms as US creators (not as a "translation step"). The +14 is not a stylistic preference; it is the empirical lift of native creative.

// For the media buyer

Surface MENA spend by register-classification (local-dialect / MSA / EN-localised) in your reporting. EN-localised cuts are silent bleeders on MENA distribution. Reallocate spend from translated cuts to native cuts before increasing budget.

// Three things to change tomorrow
01
Stop dubbing US/EU cuts for MENA distribution
The EN-localised cohort floors at 54 composite. Any spend on dubbed cuts is spend at the structural floor of the MENA market.
02
Brief in the right Arabic, not Arabic
Darija for Maghreb, Egyptian for pan-MENA reach, Khaleeji for Gulf. Choosing wrong is the largest single language-strategy failure in our cohort.
03
Treat cultural-ritual hooks as creative material, not as flavour
Iftar, family table, souk opening, prayer-call sonic, these are the strongest first-frame archetypes in the cohort. Brief them in, do not dress them on.

08 · Method.

Method. 296 ads sourced from Meta Ad Library, TikTok Creative Center, TikTok Ad Library, Google Ads Transparency Center, and manual editorial curation between 2025-Q4 and 2026-Q2 capture, across six MENA markets (MA, DZ, TN, EG, KSA, UAE). Each scored through the RoastIQ pipeline against benchmark pool v.2026-05 (n=2,047) via Vertex AI Gemini 2.5 Flash/Pro multimodal + Google Video Intelligence (shot/label) + Speech-to-Text (Arabic dialects supported via Google STT MSA + dialect models) + TranSalNet-class saliency, with Zod-validated structured output. Language strategy classification (local-dialect / MSA / EN-localised) is detected via Speech-to-Text dialect ID and manual annotation on a 90-cut training set (Cohen's κ 0.88). Confidence label MEDIUM, n_cohort < 350 per language sub-band, direction stable across pool versions, magnitudes within ±2-point bands. 5 KPIs: Beat the Skip (25%) · Get Noticed (20%) · Brand Impact (20%) · Sell Proposition (20%) · Build Brand (15%). We do not scrape. Author is Moroccan-based; the report has been internally reviewed for dialect-register accuracy.

What we do not claim. The MENA findings are model predictions calibrated against public engagement signals on the local-platform sub-pools. They do not measure in-market sales, ROAS, attributed conversion or brand recall. Confidence is MEDIUM because per-language sub-cohorts are under the 500-ad threshold for HIGH. The benchmark is still building; we expect to ship a HIGH-confidence MENA report once the cohort crosses 800 ads.

FAQ.

Why do Arabic-first ads outperform EN-localised in MENA?
Arabic-first cuts carry +14 composite points over EN-localised across our 296-ad MENA cohort. The lift is concentrated in Beat the Skip (+11) and Build Brand (+18). Local register, dialect-appropriate cadence, and culturally native hook archetypes (iftar, family table, souk opening, prayer-call sonic) earn the attention that translated copy cannot.
What is the difference between Darija, MSA and Khaleeji?
Darija is the Maghrebi dialect of Morocco, Algeria and Tunisia, code-switched freely with French. MSA (Modern Standard Arabic) is the pan-Arab written and formal register, neutral across countries. Khaleeji is the Gulf Arabic dialect of Saudi Arabia, UAE and Kuwait. The three are not interchangeable; choosing wrong is the largest single language-strategy failure in our cohort.
Which dialect wins TikTok in Morocco?
Darija dominates the Moroccan TikTok cohort with a +1.7σ composite lift over MSA and a +2.6σ lift over EN-localised. The dialect is informal, code-switches freely with French and standard Arabic, and rewards conversational creator-direct delivery. The pattern is even sharper on Algerian Darija, slightly softer on Tunisian Darija/French code-switch.
Why does MENA scroll slower than the global median?
Median MENA scroll-stop is 2.4 seconds vs 1.9 seconds globally. Three reasons: cultural-ritual hook archetypes earn slower scroll-through, lower ad saturation per session reduces pattern-recognition fatigue, and a longer-form storytelling tradition (Egyptian cinema, Khaleeji drama, Moroccan music-video culture) biases viewer expectation toward setup.
Does the UAE punish English-only cuts?
No, uniquely among MENA markets. The expat-heavy audience and regional-hub positioning make English a legitimate native register for many UAE brands. Khaleeji and MSA cuts still carry a +8 composite lift; multi-register campaigns (Khaleeji + English) outperform single-register by 4-6 points. Saudi punishes English-only the hardest in the region.
What does this mean for global brands entering MENA?
Localised copy is not localised creative. Brands that translate their global cut into Arabic and call it a MENA campaign lose the +14 lift by default. The brands winning MENA, Inwi in Morocco, STC in Saudi, Anghami across the region, build cuts in dialect, with local cultural ritual as the hook, from the strategic brief.
What is the confidence label on this report?
MEDIUM. n_cohort = 296 across six markets, but per-language sub-cohorts are under 350. The direction is stable across pool versions; magnitudes are within ±2-point bands. We expect to ship a HIGH-confidence MENA report once the benchmark crosses 800 ads. Language classification validated at Cohen's κ 0.88 against a 90-cut annotation set.

Sources. All ads in the underlying pool are sourced from official public transparency tools (Meta Ad Library API, TikTok Creative Center, TikTok Ad Library, Google Ads Transparency Center) or manual editorial curation with documented licence terms. We do not scrape. Pipeline. Vertex AI Gemini 2.5 Flash/Pro (multimodal) + Google Video Intelligence (shot/label) + Speech-to-Text (Arabic MSA + Egyptian + Khaleeji + Maghrebi dialect models) + TranSalNet-class saliency, all scored against benchmark pool v.2026-05 (n=2,047) via Zod-validated structured output. Validation. Held-out OOS Spearman ρ +0.31 TikTok engagement (n=700), +0.30 TikTok CTR (n=691), +0.32 YouTube view counts (n=403). Construct. We predict engagement and click intent against public behavioural outcomes, not sales, ROAS, attributed conversion or brand recall. Reproducibility. Every chart is pinned to model_version and benchmark_pool_version.