The playbook for landing pages that convert in Morocco in 2026.
Numbers pulled from 280 landing pages audited and optimised by Webotic for Moroccan advertisers between 2024 and Q1 2026. Core Web Vitals, hero structure, forms, trust signals, A/B testing — what actually moves conversion rate here, not in Paris or London.
Why French pages don't convert in Morocco — measured, not theorised
The median conversion rate of a Moroccan paid landing page sits at 3.2% across 280 Webotic audits between Q1 2024 and Q1 2026, against 5.8% in France and 4.6% globally. The gap isn't romantic-cultural; it's measurable, technical and editorial. First driver: digital trust. 58% of Moroccan internet users say they are wary of an online offer with no local signal, against 23% in France. A page without a recognised Moroccan client logo, without a named testimonial (Karim from Casablanca, Salma from Tangier), without aggregated Google My Business reviews mechanically converts half as well. Second driver: channel. WhatsApp covers 94% of connected Moroccans, email 67%, landline 12%. A page that offers only a form ignores the channel its audience actually uses. Third driver: French agency copy. "Discover our exclusive offer" and Parisian fashion-house lines do not resonate here. The Moroccan market rewards concrete benefit, posted price, immediate availability. Fourth and most brutal driver: speed. On Moroccan urban 4G (25 Mbps median Casablanca-Rabat, far less in semi-rural areas), a page that takes more than three seconds loses 42% of mobile visitors before the hero even paints. A page imported from an unoptimised French WordPress theme loads in a median 4.8s on Moroccan 4G — half the media budget is burned before the first scroll.
- Median conv. MA 3.2% vs FR 5.8% vs world 4.6% — 45% gap to recover.
- WhatsApp 94% vs email 67% vs landline 12% among connected Moroccans.
- Page loading > 3s on local 4G: 42% of visitors lost before hero paints.
- Adapted pages (local proof + WhatsApp + < 2s): 2.3x better than imported pages.
Core Web Vitals: the Webotic thresholds, measured on urban 4G, not in a lab
Google publishes its Core Web Vitals thresholds: LCP < 2.5s, INP < 200ms, CLS < 0.1. Those numbers are calibrated for a global median market. For Morocco we apply stricter thresholds because the speed-to-conversion correlation is harsher here: LCP < 2s, INP < 200ms, CLS < 0.1, measured on Moto G Power 2022 over urban 4G in Casablanca-Rabat, not in a controlled environment. Three levers pull LCP down. First: a hero image in WebP or AVIF under 100 KB, served with responsive srcset, never an autoplay video that consumes 800 KB before first paint. Second: critical CSS inlined in the head, the rest deferred via preload. Third: a CDN with a North-African edge — Cloudflare or AWS CloudFront via Marseille/Madrid drops TTFB from 480 ms to 180 ms on 90% of Moroccan traffic. INP, which replaced FID in March 2024, measures responsiveness to tap or click. Under 200 ms the user perceives no latency; over 500 ms the page feels sluggish and the visitor drops. To stay under 200 ms, third-party JavaScript (GTM, Meta Pixel, Insight Tag) must load deferred or async, and any heavy click handler should delegate work to a Web Worker. CLS, finally, punishes layout shifts: reserve exact image dimensions, render cookie banners as non-shifting overlays, set slider heights. A Webotic-cleared page scores 92+ on mobile PageSpeed Insights in a Morocco-4G test environment.
- LCP < 2s: hero WebP/AVIF < 100 KB, critical CSS inline, CDN edge Marseille/Madrid.
- INP < 200ms: third-party JS deferred, heavy handlers in Web Worker.
- CLS < 0.1: reserved image dimensions, cookie banner as non-shifting overlay.
- Real-world measurement Moto G Power 2022 + Orange 4G Casablanca, not Lighthouse desktop.
Hero structure: problem, solution, proof — three sentences, no more
The hero is the upper half of the page, the only zone visible before scroll. Of the 280 pages we audited, 72% lose their reader inside the hero through overpromise or under-specificity. The three-beat structure that converts best in Morocco runs as follows. First beat, the problem: a sentence that names the friction the visitor is living ("You generate Meta leads but 60% of them never pick up the phone"). This replaces the empty slogan. Second beat, the solution: a sentence that states the concrete promise with a tangible number ("We deliver 80 phone-qualified BANT+ leads in 30 days at a contracted 45 MAD CPL"). A number in the headline lifts CTR by 17% on Meta and 12% in organic search. Third beat, immediate proof: a recognised Moroccan client logo visible inside the hero, a reassurance number ("28 Meta accounts piloted since 2019"), or a short testimonial with a face. The hero should never exceed eight seconds of reading. Beyond that the user scrolls or closes — they do not read a fourth intro paragraph. For darija-speaking audiences, keep the main headline in formal French (signals credibility and purchasing power) but allow the subhead or CTA to mix in a darija touch ("Khoud rendez-vous", "Bezzaf de leads") that creates proximity. Webotic A/B tests show this French-darija mix converts 28% better than French alone on popular audiences, but 18% worse on B2B executive audiences — match the register to the audience.
- Three sentences in the hero: named problem, quantified solution, visible proof.
- Number in the headline: +17% CTR on Meta, +12% CTR organic.
- Moroccan client logo + named testimonial in the hero: social proof in under a second.
- Formal French for B2B executives; French + darija touch for popular B2C.
Forms: 3 fields for lead-gen, 4 if qualification is mandatory
Form length is the variable most directly tied to conversion rate. Across 280 tested pages, each field added beyond three drops conversion by 11% on average on a Moroccan audience, against 8% in France and 6% in the United States. For a pure lead-gen page (capture a contact to qualify downstream), three fields are enough: first name, phone, email. Phone is the pivot field: 72% reply rate in Morocco against 35% for email. Never make it optional. For a qualification page (warm sales, B2B, quote request), add a fourth field: budget, sector, or desired date. Beyond four, you lose more than you qualify — push qualification post-submission over SMS or WhatsApp instead. Client-side validation is non-negotiable. Moroccan phone: force +212 prefix, mask to 9 digits after the country code, reject 10-digit numbers starting with 0 (a common typo). Email: strict regex, reject common misspellings (gmial, hormail). The submit button should carry the action verb ("Get my audit" rather than "Submit"), sit above the mobile fold, and trigger a visible loading state. A multi-step form (2-3 short screens) outperforms a monolithic one by 59% on average, because each step asks little and progressive engagement boosts completion. Webotic recommends Typeform embed for complex multi-step forms and a native validated HTML form for short ones — Typeform adds 220 KB of JavaScript, which is best avoided if the page is already tight on performance.
- 3 fields for lead-gen (first name, phone, email); 4 if qualification is mandatory.
- Force +212 prefix, 9-digit mask, phone never optional — 72% reply rate.
- Multi-step 2-3 screens: +59% vs monolithic form on average.
- Submit button with concrete action verb, visible loading state, above mobile fold.
Trust signals: GMB reviews, client logos, certifications — where they go
In Morocco, social proof weighs more than price in the purchase decision for B2B and B2C services above 2,000 MAD. Three families of signals actually work. First family, Google My Business reviews: a "4.8/5 · 127 reviews" aggregate placed above the fold or pinned to the form lifts conversion 18 to 24%. Webotic recommends showing the true score, never rounded up — sincerity signals sincerity. With fewer than 30 reviews, drop the aggregate and show 2 or 3 named reviews with photos. Second family, recognised Moroccan client logos: three to six logos under the hero are enough. There is no point lining up twenty logos of unknown companies. If you work with OCP, ONCF, Cosumar, Wafa Assurance, Marjane or BMCE, one credible logo beats twenty illegible ones. Third family, certifications and contractual guarantees: Meta Business Partner badge, Google Partner, HubSpot certification, ISO 9001, law 09-08 CNDP compliance, legal notice reachable in one click. For high-value services, a visible "result-or-refund" guarantee — converted into a contractual clause — lifts conversion 14% on cautious B2B audiences. Absolutely avoid: unverifiable "As seen on" media badges, animated "10,000 satisfied customers" counters, mannequin stock photos. The Moroccan target decodes the artifice in under five seconds and trust collapses. Prefer three true, verifiable proofs over ten cosmetic ones.
- GMB aggregate "4.8/5 · 127 reviews": +18 to +24% conversion.
- 3 to 6 recognised Moroccan client logos under the hero, never 20 unknown logos.
- Meta/Google Partner certifications + CNDP law 09-08 compliance below hero.
- Ban stock photos, animated counters, unverifiable media badges.
CTA and placement: where, how often, with which words
The call-to-action is rarely a design problem — it is a copy and frequency problem. A well-placed CTA follows three rules. Rule one, the verb: "Get my free audit" converts 22% better than "Submit" because it names the action and the value received. "Book", "Receive", "Claim" work; "Submit", "Continue", "Validate" are neutral at best. Rule two, the frequency: one CTA in the hero (above the fold), one CTA after the social proof section, one CTA after the pricing or case study section, one CTA just before the footer. Four identical CTAs, no more. Beyond five CTAs you trigger a harassment effect that erodes trust. Rule three, the floating WhatsApp button. Across 34 Webotic A/B tests, a floating WhatsApp button bottom-right, persistently visible after the first scroll, lifts global conversion 35 to 45%. The button opens a WhatsApp Business conversation with a pre-filled message carrying the campaign source (utm_campaign) for server-side tracking. WhatsApp lead qualification rate hits 55% against 35% for plain forms — conversation filters tyre-kickers naturally. Median Webotic WhatsApp CPL: 28 MAD vs 52 MAD for a plain form. WhatsApp does not replace the form; it complements it. Pages that offer both channels convert 1.8x better than those offering only one.
- Concrete action verb: "Get my audit" +22% vs "Submit".
- 4 identical CTAs maximum per page: hero, post-proof, post-pricing, final.
- Floating WhatsApp button: +35 to +45% conversion, CPL 28 MAD vs 52 MAD form.
- WhatsApp + form combined: 1.8x better than a single channel.
A/B testing: how many variants, over which window, at which significance threshold
A/B testing is widely mispractised in Morocco, mostly through statistical under-volume. Testing two variants on 500 monthly visitors yields no reliable conclusion: noise dominates signal. Webotic applies three rules. Rule one, the minimum volume: 1,000 conversions per variant to conclude at 95% significance. Below that, extend the test or raise the media budget. On a page converting at 3%, that means 33,000 visitors per variant — roughly 6 to 8 weeks on a 25,000 MAD/month campaign. Rule two, one variable at a time: changing headline, CTA and button colour simultaneously never isolates the winning factor. Test the headline alone for 4 to 6 weeks, keep the winner, then test the CTA, then social proof. Scientific discipline pays. Rule three, the number of variants: two is enough (A vs B). Three variants dilute volume, extend duration, complicate reading. If you have three hypotheses, run three consecutive tests, not one three-armed test. Recommended tools: Google Optimize was killed in September 2023; the official successor is Optimizely Web Experimentation (expensive, 39,000 USD/year floor) or VWO (more accessible at 5,000 USD/year). For smaller accounts, GA4 Audiences combined with conditional Cloudflare Workers redirection allows a clean homemade A/B test for 0 MAD in tools — that's what Webotic runs on 80% of client accounts. Always measure downstream conversion (qualified lead, booked meeting, sale), never form clicks alone: a test that lifts submissions but degrades qualification is a losing test.
- 1,000 conversions per variant minimum for 95% significance.
- One variable at a time: headline, then CTA, then proof — never simultaneous.
- 2 variants maximum (A vs B); 3 consecutive tests beat one three-armed test.
- Measure downstream (qualified lead, booking, sale), not form clicks alone.
FREQUENTLY ASKED
- How many fields should a Moroccan landing page form have?
- Three fields for a pure lead-gen page (first name, phone, email), four if you absolutely need qualification (budget, sector, desired date). Each extra field beyond three costs 11% of conversion rate on a Moroccan audience, measured across 280 Webotic pages. Phone is the pivot field — 72% reply rate against 35% for email — and must never be optional. Beyond four fields, push qualification post-submission via SMS or WhatsApp instead of bloating the form.
- Should the landing page be in darija or French in Morocco?
- Keep the main headline in formal French for credibility and purchasing-power signal — unless you exclusively target a popular B2C audience. On subheads and CTAs, a darija touch ("Khoud rendez-vous", "Bezzaf de leads") converts 28% better on popular audiences but 18% worse on B2B executive audiences. Match the register: B2B and premium services in pure French; popular B2C in French with a darija touch; neighbourhood retail in darija with French numbers.
- Are Core Web Vitals really necessary to convert in Morocco?
- Yes, even more than in France. On Moroccan urban 4G (25 Mbps median Casablanca-Rabat, with hour-by-hour swings), a page that takes more than three seconds loses 42% of visitors before the hero paints — half the media budget burned with nothing on screen. Webotic targets LCP < 2s, INP < 200ms, CLS < 0.1 on a Moto G Power 2022 over Orange 4G Casablanca, not in Lighthouse desktop. It is demanding but directly tied to 25-35% conversion recovery.
- How many A/B variants should you test on a landing page?
- Two variants only, A vs B, never three or more. Three variants dilute the statistical volume, extend duration and complicate the reading. The minimum volume to conclude at 95% significance is 1,000 conversions per variant — roughly 6 to 8 weeks on a 25,000 MAD/month campaign for a page converting at 3%. If you have three hypotheses, run three consecutive tests (headline, then CTA, then proof), not one three-armed test. One variable at a time.
- What is the ideal length of a Moroccan landing page?
- It depends on the product's value. For a service under 2,000 MAD or a low-funnel lead, a short page (hero + 2 sections + form) is enough — around 600 words and 3 mobile screens. For a service above 10,000 MAD or a B2B decision involving multiple stakeholders, a long page (1,500-2,500 words, 6-8 sections, repeated CTAs) converts 1.4x better over 90 days because the visitor needs more proof before committing. The pivot format sits around 1,200 words.
- Does a WhatsApp button really lift conversion?
- Significantly and reproducibly. Across 34 Webotic A/B tests, a floating WhatsApp button bottom-right lifts global conversion 35 to 45%. WhatsApp lead qualification rate hits 55% against 35% for plain forms — conversation naturally filters tyre-kickers. Median Webotic WhatsApp CPL: 28 MAD vs 52 MAD for a plain form. WhatsApp does not replace the form: pages offering both channels convert 1.8x better than those offering only one.
- Should you display prices on a Moroccan landing page?
- Yes, at minimum a bracket. The absence of a price is read in Morocco as a negative signal ("If they don't show the price, it must be too expensive"). Pages that show a price or a "from X MAD" range convert 22% better than those hiding the price behind a form. For complex B2B or real estate services, display at least the budget bracket and payment terms (instalments, credit). Price transparency is a major trust factor in Moroccan commercial culture.