Core Web Vitals, in plain terms
Core Web Vitals are three metrics Google uses to measure the real experience of a page: display speed (LCP), responsiveness (INP) and visual stability (CLS). They influence ranking and, above all, conversion rate. In the Senegalese market where 3G and unstable 4G dominate, a good score is a direct commercial advantage.
This guide gives the exact thresholds, the most common causes of poor scores, concrete fixes and expected gains.
The three thresholds to know
Google classifies each metric as good, needs improvement, or poor. Systematically aim for the good zone.
| Metric | Good | Needs improvement | Poor |
|---|---|---|---|
| LCP | under 2.5 s | 2.5 to 4.0 s | over 4.0 s |
| INP | under 200 ms | 200 to 500 ms | over 500 ms |
| CLS | under 0.1 | 0.1 to 0.25 | over 0.25 |
What each metric measures
- LCP, Largest Contentful Paint: time to display the largest visible element, often the hero image or a heading.
- INP, Interaction to Next Paint: delay between a user interaction and the visual response.
- CLS, Cumulative Layout Shift: magnitude of layout jumps during loading.
LCP: causes and fixes
| Common cause | Fix | Expected gain |
|---|---|---|
| Unoptimized hero image | WebP or AVIF, exact dimensions | 0.5 to 1.5 s |
| No LCP preload | Preload the main image | 0.3 to 0.8 s |
| Slow server (high TTFB) | Cache and CDN | 0.4 to 1.0 s |
| Render-blocking fonts | font-display swap, preload | 0.2 to 0.6 s |
| Render-blocking JavaScript | Defer or remove unused JS | 0.3 to 1.2 s |
INP: causes and fixes
INP replaced FID in 2024. It penalizes pages whose main thread is saturated.
| Common cause | Fix | Expected gain |
|---|---|---|
| Too much JavaScript on load | Reduce and split the bundle | 50 to 200 ms |
| Long tasks on the main thread | Break into short tasks | 40 to 150 ms |
| Heavy event handlers | Debounce, offload to background | 30 to 120 ms |
| Heavy hydration | Partial or deferred hydration | 50 to 180 ms |
CLS: causes and fixes
| Common cause | Fix | Expected gain |
|---|---|---|
| Images without dimensions | Set width and height | 0.05 to 0.2 |
| Web fonts shifting text | Reserve space, font-display | 0.02 to 0.1 |
| Banners injected late | Reserve space upfront | 0.05 to 0.15 |
| Content inserted above the fold | Avoid high dynamic insertion | 0.05 to 0.2 |
Measuring correctly: 3G and 4G
Measuring on fiber gives a false sense of speed. The truth is in the real conditions of your visitors.
| Condition | Method | What to watch |
|---|---|---|
| Field data | CrUX report, real data | LCP, INP, CLS at the 75th percentile |
| Lab 4G | Audit with 4G throttling | LCP, total blocking |
| Lab 3G | Audit with slow 3G throttling | Time to interactive |
Need a professional website?
Kolonell builds websites that attract clients, optimized for the Sénégalese market. Free quote in 2 minutes.
The 75th percentile, not the average
Google evaluates at the 75th percentile: 75 percent of your visitors must be in the good zone. Optimizing for the average visitor is not enough; target slow connections.
Numeric example
Product page of a shop: before optimization, LCP 4.3 s, INP 320 ms, CLS 0.22 on 4G. After work: hero image switched to AVIF with preload (1.2 s gain), JavaScript bundle cut from 380 to 140 KB (180 ms INP gain), fixed dimensions on all images (CLS from 0.22 to 0.04). Result: LCP 2.3 s, INP 180 ms, CLS 0.04. All three metrics enter the good zone. Mobile bounce rate dropped 19 percent over the following month.
Prioritized action plan
| Priority | Action | Impact |
|---|---|---|
| 1 | Optimize and preload the LCP image | strong on LCP |
| 2 | Reduce JavaScript on load | strong on INP |
| 3 | Fix dimensions of all media | strong on CLS |
| 4 | Enable cache and CDN | medium on LCP |
| 5 | Optimize fonts | medium on LCP and CLS |
FAQ
Which metric should I optimize first?
LCP, almost always, because it is the most visible and the simplest to fix via the hero image. Then INP, then CLS.
Did INP replace FID?
Yes, since 2024 INP is the official responsiveness metric. It is stricter because it measures all interactions, not just the first.
Why is my lab score good but field score poor?
Because the lab often tests on a good connection. The field reflects your real visitors, often on 3G or unstable 4G. Trust the field data.
What is the acceptable CLS threshold?
Under 0.1 at the 75th percentile. The most common cause of poor CLS remains images without dimensions and injected banners.
How long before it affects ranking?
Field data updates on a rolling window of about 28 days. Expect four to eight weeks to see the impact stabilize.
Let's talk about your project. Kolonell optimizes Core Web Vitals into the green zone, with measurements to prove it. WhatsApp +221 77 596 93 33.
Mohamed Bah
Fondateur, Kolonell
Passionate about digital and entrepreneurship in Africa, Mohamed has been helping Sénégalese businesses with their digital transformation since 2020. Founder of Kolonell, he believes every SME deserves a professional and accessible online présence.
