The verdict in three sentences
A B2B website with a 480 ms INP on mobile responds poorly to every tap, and the contact and quote pages suffer most. For a New York mid-size company, a realistic 2026 budget is USD 8,000 to 30,000 in total, split between a USD 2,500 to 5,000 audit and the optimization work, delivered in 3 to 6 weeks (in France the same work costs EUR 5,500 to 18,000). Results on comparable sites range from +12 to +25% form conversion, which pays back the project within a few months once the average deal exceeds USD 5,000.
Where your site stands against Google's thresholds
Google evaluates three metrics on real visitor data (the CrUX report), at the 75th percentile. A site is rated "good" only if 75% of visits pass the threshold.
| Metric | What it measures | "Good" threshold | Typical unoptimized B2B site |
|---|---|---|---|
| LCP (Largest Contentful Paint) | Main content rendering | ≤ 2.5 s | 3.8 to 5.2 s on mobile |
| INP (Interaction to Next Paint) | Responsiveness to taps and typing | ≤ 200 ms | 350 to 600 ms, here 480 ms |
| CLS (Cumulative Layout Shift) | Visual stability | ≤ 0.1 | 0.15 to 0.30 with banners and fonts |
| TTFB (diagnostic metric) | Server response time | ≤ 0.8 s | 1.1 to 1.6 s on shared hosting |
| Third-party JavaScript weight | Chat, tags, maps, videos | < 300 KB | 700 KB to 1.4 MB |
| Mobile share of traffic | Audience exposed to the issue | n/a | 62% in our case |
INP is the most damaging metric in B2B: a quote form that freezes for half a second on each field feels broken. On mobile, that is enough for a busy buyer to leave.
Workstreams ranked by cost-to-gain ratio
Not all fixes are equal. The audit exists to prioritize those that move the 75th percentile at the lowest cost.
| Workstream | Indicative cost (USD) | Duration | Main effect | Cost-to-gain |
|---|---|---|---|---|
| Images: AVIF/WebP, sizing, lazy loading | 1,200 to 3,000 | 3 to 5 days | LCP -0.8 to -1.5 s | Excellent |
| Third-party JavaScript: audit, deferral, facades | 2,000 to 5,500 | 1 week | INP -150 to -250 ms | Excellent |
| Fonts: subsetting, preload, font-display | 600 to 1,400 | 1 to 2 days | CLS -0.05 to -0.1 | Very good |
| Cache and CDN (Cloudflare, Fastly, Bunny) | 1,000 to 3,000 + 20 to 250/month | 2 to 4 days | TTFB under 0.5 s | Very good |
| Splitting application JavaScript | 3,000 to 8,000 | 1 to 2 weeks | INP -100 to -200 ms | Good |
| Theme rebuild or move to a modern framework | 12,000 to 40,000 | 6 to 12 weeks | All metrics | Only for end-of-life sites |
On most WordPress or Drupal B2B sites, the first three workstreams are enough to get 75% of visits under 2.5 s LCP and under 200 ms INP. A rebuild only makes sense when the theme blocks any optimization.
Measure lead impact, not just the score
A Lighthouse score of 95 does not sign contracts. Tracking must connect performance to sales requests: form submission rate by device, abandonment per field, time to first click on "Request a quote". Allow 4 weeks of CrUX data after release before judging, because the report is a rolling 28-day window.
Also set a guardrail: a performance budget in the deployment pipeline (for example a maximum of 250 KB of JavaScript per page) and an alert if INP rises above 200 ms. Without it, a new marketing tool wipes out the gains within months.
Mini case study
Need a professional website?
Kolonell builds websites that attract clients, optimized for the Sénégalese market. Free quote in 2 minutes.
Michael, VP of sales at an industrial services company in New York, gets 9,000 visits a month, 62% on mobile, with a 1.4% form conversion rate, so 126 requests. Audit at USD 3,000, then fixes at USD 12,000 (images, third-party scripts, fonts, CDN): total USD 15,000. After 6 weeks, INP drops to 170 ms and conversion reaches 1.65% (+18%), so 23 more requests a month. With a 10% close rate and an average deal of USD 9,000, that is USD 20,700 in extra monthly revenue, about USD 6,200 of margin at 30%. Payback in under 3 months.
FAQ
Do Core Web Vitals really affect rankings?
They are part of the page experience signals, with modest weight compared to content relevance. The main effect is on conversion: each second of mobile LCP saved often improves submission rates by 5 to 10%.
How long before results show in Search Console?
The report uses a rolling 28 days of real data. Allow 4 to 6 weeks after release to see 75% of URLs turn green.
Should we remove live chat to fix INP?
Rarely. Loading it on first scroll or behind a clickable facade cuts 150 to 300 KB without losing conversations.
Is more expensive hosting enough?
No, the server mostly affects TTFB. Moving from USD 10/month shared hosting to a well-configured CDN at USD 20 to 50/month helps, but INP depends on JavaScript running in the browser.
Do we need a monitoring contract afterwards?
Monitoring at USD 250 to 600 per month covers metric tracking and review of every new tool. That is cheaper than a yearly audit.
Let's scope your project. We audit your site, price each fix by cost-to-gain and set a measurable INP and lead target, usually within 3 to 6 weeks. Detailed quote within 48 h. 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.
