Góc Nhìn Mới: Tại Sao Core Web Vitals Và Chuyển Đổi Không Nên Bắt Đầu Bằng Công Cụ?

Picture a meeting. On one side is your Dev, PageSpeed Insights report in hand, beaming with confidence: "98 out of 100, boss - all green!" On the other side is your Head of Sales, revenue report in hand, looking... less than thrilled: "Great score, but this month's orders are exactly the same as last month. Stuck in place."

So who's right? Both of them. And that's exactly the problem.

The tool isn't lying - it just isn't telling the whole story

Core Web Vitals - the trio of metrics made up of LCP (how fast the main content loads), INP (how responsive the page is to interaction), and CLS (how visually stable the layout stays) - are the standard Google itself uses to evaluate page-loading experience. The issue isn't that these metrics are wrong. The issue is how many agencies and businesses have turned them into the final destination, instead of treating them as a necessary condition.

There's a line worth sitting with from SEO expert Dan Taylor, published on Search Engine Land in early 2026: Core Web Vitals should be understood as a "gate," not a "signal of excellence." What does that mean in practice? Getting through that gate (hitting a "Good" score) is the minimum bar to even be considered - but moving your score from "good" to "great" doesn't automatically translate into proportionally better business results.

This is exactly why the relationship between Core Web Vitals and conversion needs to be rethought - not by "scoring the tool first, worrying about conversion later," but the other way around.

So what does "starting in the right place" actually mean?

Back to that meeting. If you start from the tool, the whole team ends up asking: "How do we get LCP under 2.5 seconds?" But if you start from conversion, the real question becomes: "Where in the journey are our customers dropping off, and is page speed actually the cause?"

The two questions sound similar, but they lead to two completely different directions. The first question keeps the team narrowly focused on patching technical issues to make the score look good. The second forces the team to look at the overall website UX - from information architecture, to mobile design, to the form-filling experience - not just load speed alone.

Real-world data shows just how concrete the results can be when you take the right approach. According to a Google case study published on web.dev, based on an analysis of over 10 million visits across 33 countries, Renault improved LCP by 1 second and saw its bounce rate drop by 14 percentage points, alongside a 13% lift in conversion rate. Vodafone reported similar results: improving LCP by 31% drove an 8% increase in sales and a 15% increase in add-to-cart rate.

What's worth noting: neither of these cases started with the question "how do we hit a perfect score." They started with "where exactly are customers running out of patience in their buying journey."

A few numbers to show you just how widespread this problem is

According to HTTP Archive's 2025 Web Almanac Report - one of the largest website data sources in the world today - only 48% of mobile websites pass all three Core Web Vitals thresholds at the "Good" level. In other words, more than half of the websites out there running ads, pouring money into marketing, are already working against themselves at the level of website speed and mobile experience - factors that should really be handled before anyone even gets to the higher-level work of optimizing website CRO.

So where should you actually start?

The answer isn't "ignore the tool." It's "reverse the order of priorities":

1. First, understand the customer's actual journey on your website - where they come from, what they're looking for, where they drop off.
2. Then, identify whether website speed, mobile design, or overall UX is the biggest bottleneck in that journey.
3. Only then open up tools like PageSpeed Insights or the Core Web Vitals report to measure and fine-tune - treating them as a diagnostic check, not a compass.

For agency websites - where customers typically go through several stages of research before making contact - getting this order of priorities right matters even more, because a single small friction point in the experience can undo all the effort put into building that content journey in the first place.

Don't let your meeting end in an argument

If your team is also stuck in the "great score, flat revenue" situation, it might be time to change the question. Reach out to UNITRUX today - we'll help you take a fresh look at the real relationship between Core Web Vitals and conversion, starting from the customer journey, not from the scoreboard.

Sources: Dan Taylor - Search Engine Land (2026), Google web.dev Case Studies (Renault, Vodafone), HTTP Archive Web Almanac 2025.