# Fixing empty PageSpeed results on Terlihat.com

Inspect the before-and-after evidence from correcting the PageSpeed filter and Google requests in Cloudflare.

[![](https://terlihat.com/images/authors/yudistira-ashadi-96.webp)Yudistira Ashadi](https://yudis.dev/about)Updated  20 November 2026 2 min read

![Hands connecting a cable at a test bench.](https://terlihat.com/images/articles/pagespeed-worker-fix-1600.webp)

On November 20, 2026, the speed comparison returned HTTP 200 with empty results for two test pages. After correcting the PageSpeed filter and adapting the requests for Cloudflare Workers, the production recheck returned Lighthouse scores for both pages and CrUX data for web.dev.

## Check what an HTTP 200 actually returned

We tested the Terlihat.com homepage and web.dev in production. The tool reported a configured Google key. Both rows lacked Lighthouse scores and real-user data. Our completion check called for two Lighthouse scores and real-user data for the test URL with coverage.

Scroll the table sideways to compare all columns.

| Evidence | Before | After |
| --- | --- | --- |
| Terlihat.com endpoint status | HTTP 200 | HTTP 200 |
| Pages with Lighthouse scores | 0 of 2 | 2 of 2 |
| Pages with CrUX data | 0 of 2 | 1 of 2 (web.dev) |

## Correct the Google field filter and redirect handling

Google rejected our filter for individual properties inside the Lighthouse audit map. We requested the map and extracted the required metrics after the response arrived. The local test then worked, but production still returned empty rows. Cloudflare Workers rejected the redirect mode before the request went out. We switched to manual handling and continued to reject redirect responses.

Source: [Cloudflare Workers -- Request redirect modes](https://developers.cloudflare.com/workers/runtime-apis/request/)

## Run the same two inputs in production

The production recheck used the same URLs. The result we recorded was the return of usable measurements. Lighthouse scores can vary between runs, and URL-level CrUX data depends on coverage. This recheck verifies the measurement tool. It provides no evidence of a change in site speed or marketing outcomes.

Source: [Chrome for Developers -- CrUX URL-level data](https://developer.chrome.com/docs/crux/api)

[Download the verification record (JSON)](https://terlihat.com/downloads/pagespeed-recheck-2026-11-20.json)[Try the mobile speed comparison](https://terlihat.com/en/mobile-speed-comparison)[Open the page audit](https://terlihat.com/en/page-audit)

## Put it into practice.

Answer four questions to identify your business's next step.

[Check AI-search readiness](https://terlihat.com/en/readiness-check)

## Keep reading.

[![](https://terlihat.com/images/articles/live-technical-audit-640.webp)A live technical audit of a Terlihat.com page](https://terlihat.com/en/articles/live-technical-audit-example)[![](https://terlihat.com/images/articles/technical-640.webp)A technical SEO checklist for your most important pages](https://terlihat.com/en/articles/technical-seo-checklist)[![](https://terlihat.com/images/articles/free-ai-visibility-tools-640.webp)Free AI visibility tools, depending on what you want to check](https://terlihat.com/en/articles/free-ai-visibility-tools)

## Start with one buyer question.

Bring a question you want to test in ChatGPT or Google. We'll discuss what to audit, which pages could improve, and what help you need.

[Tell us about your business](https://terlihat.com/en/contact)

---

Source: [Terlihat.com](https://terlihat.com/en/articles/fixing-pagespeed-on-cloudflare)
