SEO BrainFree · Built in Bahrain
Book a DemoLet’s Connect

Module M0

A technical check you can repeat every week

One page or template, checked against the same list every time, with three performance runs instead of one.

10 min readIntermediate80 XP quest
Rows of server cabinets in a large data centre
Photo: Freepik via Magnific (licensed stock)

Core Web Vitals: where "good" ends

Google's thresholds at the 75th percentile of real visits, unchanged as of November 2026 (web.dev). The needle marks the good limit.

LCPgood ≤ 2.5 s · poor > 4 s
INPgood ≤ 200 ms · poor > 500 ms
CLSgood ≤ 0.1 · poor > 0.25

The checklist

Indexing

  • Status 200, indexable, self canonical
  • In the sitemap, not blocked by robots.txt
  • No redirect chains, no soft 404s

On-page

  • Title 30 to 60 characters, unique
  • Description 70 to 160 characters
  • One visible H1 (a 1 px clipped H1 reads as hidden text)
  • First 100 words say what, for whom, where

Structured data

  • Only schema that matches visible content
  • Organization with legal identifiers
  • No review stars without genuine, eligible reviews

Performance

  • Three comparable Lighthouse runs, median and spread
  • Lab data is not field data
  • Targets: LCP 2.5 s, CLS 0.1, INP 200 ms

Why three runs

On our own homepage, runs on the same day varied by up to 18 performance points [T] 25 Oct 2026. A single run would have sent us chasing noise.

BreakdownStep by step

  1. Fetch the page like a crawler

    Status code, redirects, the raw HTML without JavaScript. If the main text is missing from the raw HTML, nothing else on this list matters yet.

  2. Check indexability

    No noindex tag or header, a self canonical, listed in the sitemap, not blocked in robots.txt.

  3. Check the words search engines read first

    Title 30 to 60 characters, a description of 70 to 160 that says what, where and the next step, one visible H1 that names the service.

  4. Check structure and links

    H2s and H3s in logical order, descriptive internal links in and out, alt text on meaningful images.

  5. Check schema against the page

    Every JSON-LD block parses and states only what is visible on the page.

  6. Measure speed properly

    Three mobile runs for LCP, CLS and INP; report the spread and keep field data apart.

Worked example

A service page check, row by row (illustrative)

Status 200, no redirect [T]. Canonical points to the page itself [T]. Title "Interactive Screen Rental Bahrain | FAIM" is 40 characters [T]. Two H1s: the hero line and a section heading [T]; fix by making the section an H2. Hero image has alt "IMG_0412" [T]; replace with "Touchscreen kiosk at an exhibition stand in Manama". LCP 2.9, 3.1, 3.0 s on three runs [T]; the hero video is the LCP element [A].

TipsWhat works in practice

Run M0 on a template, not every page. Fixing the template fixes all its pages.
Check the mobile view at 390 pixels wide; most buyers in the Gulf search on phones.
Read the page title out loud. If it would sound odd on the phone, rewrite it.
When a page builder adds a second H1, fix the tag and keep the look with CSS.

AvoidCommon mistakes

  • Hiding a heading to pass the one-H1 check.
  • Adding schema that describes things the page does not show.
  • Fixing speed before indexability.

Quest · Technician badge

Inspect a page

3 tasks + 1 question · 80 XP

The main text is missing from the raw HTML but appears in the browser. What does that mean?

Why: Many AI crawlers do not run JavaScript, and Google renders it later. Content should be in the HTML that arrives first.

Technician badge unlockedChapter complete. Your progress is saved in this browser.