---
title: "What Are Core Web Vitals? Speed Metrics for an E-Commerce Site (LCP, INP, CLS)"
description: "What are Core Web Vitals? Learn what LCP, INP and CLS measure, typical problems on e-commerce pages, solutions and the order of improvements."
url: https://www.faprika.com/en/blog/what-are-core-web-vitals
language: en
site: Faprika
---

# What Are Core Web Vitals? Speed Metrics for an E-Commerce Site (LCP, INP, CLS)

> What are Core Web Vitals? Learn what LCP, INP and CLS measure, typical problems on e-commerce pages, solutions and the order of improvements.

A visitor opens a product page in your store. The page arrives but the product image stays blank for a few seconds; just as the visitor is about to press “Add to Cart”, a section above loads and pushes the button down, so the tap lands in the wrong place; and when they do press the button, the page freezes for a moment. None of these is a problem you'd describe as “the site didn't open”, but each is reason enough to abandon the purchase. Search engines now measure this experience too and report it as three separate metrics.

In this guide we answer the question **what are Core Web Vitals** from the point of view of an e-commerce store: what the three metrics (LCP, INP, CLS) mean, how to read the results, the most common problems and solutions on product, category and home pages, the difference between field data and lab data, the order in which to make improvements, and a checklist. At the end you will find information about Faprika and a six-question FAQ.

Key Takeaway

“Instead of reducing page speed to a single ‘loaded’ moment, Core Web Vitals measures it with three questions: when did the content appear, how quickly did the page respond to a tap, and did what you see stay in place? In e-commerce these three questions map directly onto the shopping flow: the product image, the add-to-cart button and the price area. Use the metrics not to chase numbers but to find the three frustrations your visitor experiences.”

## What Are Core Web Vitals?

Core Web Vitals are a set of measurements made up of three indicators that measure how fast a page appears to real visitors, how quickly it responds and how stable it stays: LCP, INP and CLS. Search engines use these indicators as part of their page experience signals.

Although the metric names are abbreviations, the idea is simple. In the past, “how long did the page take to load?” was answered with a single time. Yet a page can be usable before it has fully loaded, or look loaded and be unusable. That is why Core Web Vitals measures not loading but the experience the visitor feels. The short meaning of the three metrics is as follows:

- **LCP (Largest Contentful Paint):** When the main content of the page, that is, the largest element visible on screen, appears.
- **INP (Interaction to Next Paint):** How quickly the page gives a visible response when a visitor taps a button, a menu or a field.
- **CLS (Cumulative Layout Shift):** How much elements on the page move around while loading.

These metrics are only one side of site speed. For the general framework of topics such as server response time, network connection and page weight, see the post [why site speed matters in e-commerce](https://www.faprika.com/en/blog/e-commerce); this post focuses on its measurement side.

*Görsel: What are Core Web Vitals: LCP measures when content appears, INP the response time to a tap, and CLS the layout shift (Three metrics measure three different moments on a product page)*

## What Do LCP, INP and CLS Measure?

LCP measures how fast content appears, INP how fast the page responds to interaction, and CLS visual stability. Each looks at a different period of the page: LCP at the moment of opening, INP at the time of use, and CLS at the shifts that occur over the page's lifetime.

### 1. LCP: when did the main content appear?

LCP is the time until the largest element visible in the first part of the screen, usually an image or a large block of text, is painted. In e-commerce this element changes with the page type: on a product page it is the main product image, on a category page the product images in the first row or the category image, and on the home page usually the large image at the top.

For LCP to count as “good”, the main content must appear shortly after the visitor opens the page. Four things delay it: a slow-responding server, a heavy image, a style or script that holds up the screen, and the first image starting to load late. In most stores the problem is the second and the fourth.

### 2. INP: how quickly did the page respond to a tap?

INP measures how quickly a visitor's clicks, taps and key presses on the page show up on screen. It looks at the interactions made over the page's lifetime and highlights the slowest one. Typical examples are the cart icon updating when “Add to Cart” is pressed, the list changing when a filter is selected, and a menu opening.

What spoils INP is usually scripts that keep the browser's main thread busy for a long time. Marketing tags running at the same time, a heavy cart script or a large filter code leaves the visitor's tap waiting in a queue. The result: the button is pressed but nothing happens, and the visitor presses again.

### 3. CLS: did what you see stay in place?

CLS is the total measure of elements moving unexpectedly while the page loads or is being used. Text shifting down while you read, or the “Add to Cart” button moving elsewhere just as you press it, are signs of it. Low is good: if nothing jumps while the page loads, the result approaches zero.

In e-commerce the main causes of shifts are images without specified dimensions, ad and notification banners that load late, font swapping, and cookie or campaign bars added to the top of the page afterwards.

| Measurement | What does it measure? | Equivalent on the product page | Direction |
| --- | --- | --- | --- |
| LCP | When the main content appears | The large product image being painted | The shorter, the better |
| INP | Time to visible response to an interaction | The reaction to add-to-cart or option selection | The shorter, the better |
| CLS | Amount of layout shift | Price and button shifting position | The lower, the better |

We recommend reading

How to Optimize Product Images?

## How to Read Core Web Vitals Results

Each metric falls into one of three bands: good, needs improvement, or poor. A page’s overall status depends on all three metrics being “good”; if one is poor, the page as a whole is considered to need improvement.

Measurement tools display the threshold values for each metric themselves; we don’t give numbers in this article because the thresholds may be updated by search engines over time, and it is more reliable to look at the band the tool shows than to memorize a number. What matters is reading the report correctly:

- **Look at the band color, not at a single number.** The transition between the good, needs improvement, and poor bands also reflects the difference the visitor feels.
- **Look at the page group.** Reports often group similar pages: all product pages, all category pages. A problem in one template affects hundreds of pages at once.
- **Read mobile and desktop separately.** The same page may be poor on mobile and good on desktop; since most shopping happens on mobile, look at mobile first.
- **Put the weakest metric first.** If one of the three is poor, fix it first; improving a metric that is already good won’t change the outcome.

| Band | Meaning | What to do? |
| --- | --- | --- |
| Good | Most visitors have a comfortable experience on that metric | Maintain it; measure again when adding new features |
| Needs improvement | The experience is acceptable but some visitors have trouble | Planned improvement; start with the product and category templates |
| Weakness | A significant share of visitors have problems | Priority fix; tackle the most problematic template |

Tip

Look for the cause, not the result

A poor measurement doesn’t mean “rewrite the site from scratch.” Often a single thing is responsible: a main image with a large file size, a few images without dimensions, or a single heavy script running on first interaction. First find which element is hurting which metric, then fix them one by one.

## What Is the Difference Between Field Data and Lab Data?

Field data is the aggregated experience of real visitors on their own devices and connections; lab data is the result of loading a page once, artificially, under assumed device and connection conditions. They answer different questions and often don’t give the same number.

This difference is what e-commerce managers confuse the most. If one tool says a page is good and another says it is poor, one is usually looking at field data and the other at lab data.

*Görsel: What are Core Web Vitals: a comparison of field data collected from real visitors and lab data produced in a controlled environment (Field and lab data answer different questions)*

### Field data: what actually happened?

Field data is collected from the devices of visitors who actually open your page. Low-end phones, slow mobile connections, different browsers, and real interactions are all taken into account. This is the data search engines rely on when evaluating page experience. Its weakness is that it doesn’t show a recent change immediately, and for pages with few visits there is often not enough data.

### Lab data: why is it like this?

Lab data loads the page under fixed conditions and records every step. It shows which element is the LCP, which script holds the page up, and what causes shifts; you can see the effect of a change immediately. Its weakness is that it doesn’t reflect real visitor diversity and can’t fully measure interaction-based metrics such as INP, because nobody actually touches the page.

| Feature | Field data | Lab data |
| --- | --- | --- |
| Source | Real visitors | Controlled artificial load |
| What does it show? | What actually happened | The cause and what is breaking it |
| Update | Builds up over time, appears late | Instant on every run |
| Low-traffic page | Data may be insufficient | Every page can be measured |
| What is it used for? | Confirm the situation, set priorities | Find the problem, try a fix |

The practical approach is this: first look at field data to determine which page group is problematic, then use lab data to find the cause and fix it, and after a few weeks check in the field data whether the result has improved. Most measurement tools present these two kinds of data in separate sections of the same report.

## What Are the Typical Problems on E-Commerce Pages?

Product, category, and home pages tend to harm different metrics: on the product page it is usually the main image and option interaction, on the category page the weight and shifts caused by many images, and on the home page the large hero image and late-loading sections.

*Görsel: What are Core Web Vitals: typical speed problems and related metrics on the home page, category page and product page (The most common problems by page type)*

### Product page: main image and selections

On a product page, the LCP is usually the main product image. If the image is larger than necessary, in an old format, or drawn late after a gallery script loads, LCP gets worse. Size, color, and quantity selections and the “Add to Cart” flow determine INP; if the script running in this flow is heavy, taps are answered late. If price, stock information, or a campaign badge loads afterward, CLS suffers.

### Category page: many images and filters

A category page contains dozens of product images. Loading all of them at once weighs the page down; when their dimensions aren’t specified, the list shifts downward as they load and CLS rises. If filter and sort options run with a heavy script, INP weakens. For product list and filter layout, the articles [product filtering and search bar optimization](https://www.faprika.com/en/blog/product-filtering-and-search-bar-optimization-in-e-commerce) and [category page SEO](https://www.faprika.com/en/blog/category-page-seo) will be useful.

### Home page: large hero and many sections

On the home page, the LCP is mostly the large image in the hero. Showcase, campaign, and slider sections stacked one after another, especially when loaded late, cause shifts lower on the page. Multiple sliding images (sliders) bring both weight and interaction cost.

| Page | Most affected metric | Typical cause |
| --- | --- | --- |
| Product page | LCP, CLS | Heavy main image, price and badge arriving late |
| Product page | INP | Heavy script in the option and cart flow |
| Category page | CLS, INP | Images without dimensions, heavy filter script |
| Home | LCP, CLS | Large hero image, late-loading sections |

## How to Improve LCP?

The way to improve LCP is to find the largest element in the first part of the screen (often an image), make it smaller, and tell the browser to load it first. The biggest gain is usually on the image side.

### 1. Bring the main image to the right size and format

Loading an image that will look small on screen at a very high resolution is the most common mistake. An image should be at most twice the size at which it is displayed and should be served in a modern, well-compressed format (e.g. WebP). In Faprika, an image uploaded to the panel is saved in three separate sizes; which screen uses the small, medium, and large sizes is explained in the training page. For a detailed guide, see the [product image optimization](https://www.faprika.com/en/blog/product-image-optimization) article.

### 2. Don’t lazy-load the first image

Lazy loading is a technique that postpones images that haven’t come near the screen and lightens the page; it is very useful for images below the fold. But if it is used on the main image in the first screen, it delays LCP, because the browser holds the image back as “not needed yet.” Give the main image priority and lazy-load the ones below.

<code><!-- İlk ekrandaki ana ürün görseli: tembel değil, öncelikli --> <img src="https://magazaadresiniz/urun-ana.webp" width="800" height="800" fetchpriority="high" alt="Beyaz spor ayakkabı, yandan görünüm"> <!-- Ekranın altındaki görsel: tembel yükleme --> <img src="https://magazaadresiniz/urun-detay.webp" width="800" height="800" loading="lazy" decoding="async" alt="Beyaz spor ayakkabı, taban detayı"></code>

### 3. Reduce the resources that block rendering

Large stylesheets and scripts placed at the top that hold up page rendering delay LCP. The styles needed for the first screen are inlined or kept small; scripts not needed for the first paint are deferred. Likewise, speeding up the server’s response and serving images through a content delivery network (CDN) improves LCP, especially for visitors coming from different regions. Faprika’s plan comparison includes a CDN row; its scope may vary by plan.

*Görsel: What are Core Web Vitals: the LCP improvement flow; finding the main image, reducing its size, prioritizing it and lazy-loading the rest (Four-step improvement flow for LCP)*

We recommend reading

E-Commerce

Product Page: What Should It Look Like?

## How to Reduce CLS?

The rule for reducing CLS is simple: the space needs of every element on the page must be known before the element loads. If space is reserved in advance for images, ads, notification bars, and fonts, the page doesn’t jump while loading.

### 1. Write width and height for every image

An image without width and height takes up zero space until it loads; when it loads, it pushes everything below it down. If the dimensions are written in the HTML, the browser reserves the correct space before the image arrives. This is the cheapest and most effective fix for CLS in e-commerce; as seen in the code example above, every image should have width and height.

### 2. Reduce shifts caused by font swapping

When a custom font loads late, the text is first drawn with a fallback font and then changes place; line lengths change and the content below shifts. Keeping the font file small, loading only the weights you need, and bringing the fallback font close to the main font in metrics reduce this shift.

<code>@font-face { font-family: "MagazaYazi"; src: url("https://magazaadresiniz/yazi.woff2") format("woff2"); font-display: swap; }</code>

### 3. Reserve space for sections that arrive later

If sections such as a campaign bar, a cookie notice, a “recently viewed” strip, and a review summary are added after the page is drawn, they push the content above or below them. Reserve a fixed height for these sections, or show them in a form that doesn’t push the page flow, overlaid on top of the screen. The same rule applies to ad and notification bars.

### 4. Watch out for late-loading sections

Loading sections below the screen late is good for speed; but if the section’s height isn’t known when it arrives, everything below it shifts. Give placeholder heights to late-loading sections. If the visitor hasn’t scrolled to that section, the shift may not count in the measurement, but jumps while scrolling still spoil the experience.

Tip

Keep the area beneath the Add to Cart button fixed

The most expensive place for a shift is the button the visitor is just about to press. Keep the height of the price, stock, and size selection areas fixed; even if product information fills in later, the button should stay in place. That way CLS drops and mis-clicks decrease.

We recommend reading

How to Write Image Alt Text?

## How to Improve INP?

Improving INP means reducing the work that keeps the browser’s main thread busy and reflecting a response on screen right after a tap. The biggest source of this is usually scripts added to the page later.

### Limit third-party scripts

Measurement, advertising, live support, review, and notification tools add third-party scripts to the page. Each of them spends time on the main thread; when all of them run at once, taps wait in a queue. Ask this about each script: is this tool actually being used, must it load on the first paint, can it load after interaction? Remove unused tags and make the remaining ones deferred or loaded after the first interaction. On Faprika’s own site, measurement and live support tools are likewise loaded on first interaction.

<code><!-- Kritik olmayan betik: sayfa çizimini beklemez --> <script src="https://magazaadresiniz/olcum.js" defer></script></code>

### Break up long tasks

If a script runs uninterrupted for a long time, the browser can’t respond to taps during that time. Splitting large jobs into small pieces or leaving them until after the first paint improves INP. For frequently used interactions such as filtering and search, showing a loading indicator right after the tap gives the visitor the message “your click was received.”

### Keep interactive elements simple

Unnecessary animation, effects that force the whole page to be recalculated on every tap, and heavy pop-ups slow down interaction. Keeping frequently used elements such as the menu, cart, and filters simple and light is the safest path.

| Problem | Affected metric | Solution |
| --- | --- | --- |
| Very large main image | LCP | Reduce the size, use a modern format |
| Lazy loading of the first image | LCP | Don’t lazy-load the main image, give it priority |
| Image without specified dimensions | CLS | Write width and height |
| Late-arriving font | CLS, LCP | Shrink the file, bring the fallback font closer |
| Bars and sections added afterward | CLS | Reserve space or show without pushing the page |
| Many third-party scripts | INP, LCP | Remove, defer, load after interaction |
| Long-running scripts | INP | Split the work, give immediately visible feedback |

## In What Order Should Improvements Be Made?

The right order is to start with the template that affects the most visitors and do the easiest and most effective fixes first. In practice, this order is the product page template, the category template, the home page, and finally the other pages.

*Görsel: What are Core Web Vitals: an eight-item Core Web Vitals checklist for an e-commerce site (Eight-step checklist for Core Web Vitals)*

1. **Measure and identify the problem page group.** Find in the field data which template is weak on which metric; look at mobile first.
2. **Fix the images.** Correct the size and format of the main image, don’t lazy-load the first image, and write width and height for all images. This step improves LCP and CLS together.
3. **Close off the sources of shifting.** Reserve space for bars and sections that arrive later, and make the font lighter.
4. **Review the scripts.** Remove unused tags and defer the rest; INP improves at this step.
5. **Measure again.** Check the result immediately with lab data; wait a few weeks in the field data.
6. **Turn protection into a rule.** Before adding a new add-on, campaign bar, or tag, measure again on product and category pages.

Tip

Measure every change one by one

If you make five changes at once, you can’t tell which one worked and which one did harm. Make changes in small groups and look at the lab data after each one. Adding a tag is also a change: repeat the same measurement whenever a new tool is added.

We recommend reading

How to Set Up

E-Commerce

SEO Settings

## What Are Common Mistakes in Core Web Vitals?

Most mistakes come from misreading the measurements and judging the whole site by looking at a single page. The most common ones are these:

- **Measuring only the home page.** A store’s traffic and revenue are on product and category pages; the problem usually shows up there.
- **Relying only on lab data.** It doesn’t show what real visitors experience; field data is essential, especially for INP.
- **Looking only at desktop.** Mobile devices run slower; problems show up there first.
- **Lazy-loading the main image.** This setting, made to lighten the page, hurts LCP.
- **Not writing dimensions for images.** It is the most common cause of CLS and the cheapest to fix.
- **Adding new tags without measuring.** Every new third-party script costs INP and LCP.
- **Chasing numbers.** Chasing a score and overlooking the problem the visitor actually experiences. The goal is not to turn the metric green but to improve the experience.

## Core Web Vitals Checklist

The list below contains eight questions that can be reviewed before putting a product or category template live and after every major change. It is the same as the list in the image.

| # | Question | Related metric |
| --- | --- | --- |
| 1 | Is the main image in the first screen the right size and in a modern format? | LCP |
| 2 | Is the main image not lazy-loaded and given priority? | LCP |
| 3 | Are the images below the screen lazy-loaded? | LCP, weight |
| 4 | Are width and height written on every image? | CLS |
| 5 | Have you reserved space for banners and sections that load later? | CLS |
| 6 | Is the font file small and the fallback font close in size? | CLS, LCP |
| 7 | Have unused third-party scripts been removed and the remaining ones deferred? | INP, LCP |
| 8 | Is field data tracked separately for mobile and desktop? | All |

Don’t apply this list just once; apply it at regular intervals. A store’s speed degrades gradually over time with campaign banners, new add-ons and new images; regular checks catch this slowdown early.

## Core Web Vitals and Site Speed with Faprika

Faprika’s plan feature comparison includes speed-related rows: “CDN Content Delivery Network” and “Lazy Load / Gzip”. The scope of these features in each plan may vary; for the current scope, see the [e-commerce plans](https://www.faprika.com/en/pricing) page. The panel also has a “Media Settings” area for images: every image you upload is saved by the system in three sizes, small, medium and large, and these sizes can be edited in the settings; if you use a CDN, you can also enter its address in this area. This way you don’t have to load the same large image in the cart preview, on the storefront and on the product detail page.

How to set up image sizes, how to fill in SEO fields and how to update meta data in bulk are covered in the “Media Settings Management” and “SEO Settings” tutorials in Faprika’s help center. Page speed is the result not only of the software but also of the images you upload, the tools you add and your design choices; so you can review together with the Faprika team which setting will change what in your store.

To set up your site’s general SEO settings together, see the articles [how to configure e-commerce SEO settings](https://www.faprika.com/en/blog/how-to-set-up-e-commerce-seo-settings), [what a URL structure should look like](https://www.faprika.com/en/blog/what-should-a-url-structure-look-like) for site addresses, and [how to write image alt text](https://www.faprika.com/en/blog/how-to-write-image-alt-text) for images.

## Frequently Asked Questions

**What are Core Web Vitals?**

Core Web Vitals are three indicators that measure how fast a page appears to real visitors, how quickly it responds and how stable it stays: LCP, INP and CLS. Search engines use these indicators as part of their page experience evaluation.

**What do LCP, INP and CLS mean?**

LCP measures when the largest content element of the page appears; INP measures how quickly the page visibly responds to a visitor's tap; and CLS measures how much elements on the page move while loading. A short result is good for LCP and INP, a low result is good for CLS.

**Do Core Web Vitals affect rankings?**

Yes, but they are not decisive on their own. Search engines use page experience as a signal; the relevance of the content to the topic is a far heavier factor. Even so, a poor experience matters in its own right, because it makes visitors leave the site and lowers conversion.

**Why do field data and lab data differ?**

Field data shows the experience real visitors have with their own devices and connections, while lab data shows the result of loading the page artificially under fixed conditions. Because they measure different conditions, they don't give the same number; field data confirms the situation, lab data helps find the cause.

**How do you improve Core Web Vitals on an e-commerce site?**

First find out which metric your most visited product and category templates are weak on. Then shrink the main image and don't lazy-load the first image, set width and height on all images, reserve space for sections that load later, remove or defer unnecessary third-party scripts, and measure again after every change.

**Why should mobile and desktop results be viewed separately?**

Mobile devices run with a more limited processor and connection; the same page can score poorly on mobile and well on desktop. Because most shopping happens on mobile, you should start improving with mobile results and track the two device types separately.

A fast store starts with its images and layout

If you are looking for an e-commerce infrastructure where you can manage image sizes, CDN and speed settings from the panel, explore Faprika’s plans. On the page you can compare which plan covers which speed and SEO features.

Explore the Plans

#### You May Also Like

- [**Why Does Site Speed Matter in E-Commerce?**](https://www.faprika.com/en/blog/e-commerce)
- [**How to Optimize Product Images?**](https://www.faprika.com/en/blog/product-image-optimization)
- [**What Should an E-Commerce Product Page Look Like?**](https://www.faprika.com/en/blog/e-commerce-product-page)
- [**How to Do Category Page SEO?**](https://www.faprika.com/en/blog/category-page-seo)
