heartex.ai website review

20 files are served with short cache lifetimes, so returning visitors download them again (746 KiB).

About 641 KiB of JavaScript is downloaded but never used on this page. Google estimates it would make the page about 3.6 s faster on a phone.

14 files block the first paint until they finish downloading. Deferring or inlining them could save about 1.3 s.

8 third-party services load with your homepage (Google Tag Manager, Facebook and Hubspot and others), adding 1.7 MB and 1.1 s of main-thread time.

Your mobile performance score is 59 (desktop: 70). Pages that score under 90 lose visitors before the first paint.

6 images could be 621 KiB smaller with modern formats (WebP or AVIF) and sizes that fit a phone screen.

Google Tag Manager costs a phone 439 ms of main-thread work and 686 KiB of downloads while your page loads. Loading it after the page is usable, or only on the pages that need it, takes that off the critical path.

Polyfills for old browsers add 71 KiB JavaScript that modern phones do not need. Google estimates it would make the page about 400 ms faster on a phone.

Facebook costs a phone 191 ms of main-thread work and 304 KiB of downloads while your page loads. Loading it after the page is usable, or only on the pages that need it, takes that off the critical path.

A phone's processor spends 2.1 s on main-thread work while loading, 1.3 s of it running JavaScript. Above 2 s, taps and scrolls feel sluggish.

Hubspot costs a phone 160 ms of main-thread work and 311 KiB of downloads while your page loads. Loading it after the page is usable, or only on the pages that need it, takes that off the critical path.

11 long tasks block the main thread while loading, the longest for 153 ms. During a long task the page ignores taps.

Critical files load one after another instead of in parallel (3 chains). Preconnecting to key origins and preloading the hero image shortens the chain.

Scripts make the browser recalculate the layout repeatedly while the page loads, which wastes processor time on phones.

Real Chrome users on phones do not pass Google's Core Web Vitals on this page (28-day field data from the Chrome UX Report (opens in a new tab)). Google uses this in ranking.

Your largest element takes 23.6 s to render on a phone. Google's threshold for “good” is 2.5 s. The largest element is the text “Labeling, evaluation, and quality workflows that deploy whe…”.

The page becomes fully interactive after 23.6 s. Good is under 3.8 s.

Visually, the page fills in at 6.0 s (Speed Index). Good is under 3.4 s.

Real visitors wait 286 ms for the page to respond to a tap (75th percentile). Good is under 200 ms.

Over the last 25 four-week periods to 2026-10-17, real Chrome visitors' tap response (inp) went from 221 ms to 274 ms at the 75th percentile (Chrome UX Report (opens in a new tab) history).

Real visitors see the first content after 1.9 s (75th percentile). Good is under 1.8 s.

Real visitors wait 919 ms for your server to start answering (75th percentile). Good is under 800 ms.

Scripts block taps and scrolling for 421 ms in total while the page loads. Good is under 200 ms.

8 images have no width and height attributes, so the layout jumps as each one loads.

Over the last 25 four-week periods to 2026-10-17, real Chrome visitors' largest content (lcp) went from 2.0 s to 2.0 s at the 75th percentile (Chrome UX Report (opens in a new tab) history).

Your homepage downloads 4.0 MB in 113 requests on a phone: images 2.1 MB (26), scripts 1.8 MB (33) and fonts 71 KiB (2). Under 2 MB keeps phones on slow connections fast.

evaluation-ui.png weighs 1.3 MB. On a phone connection a single image that size takes seconds.

9 images (2.1 MB) are JPEG, PNG or GIF. The same pictures as WebP or AVIF are usually much smaller at the same quality, and every current browser shows them.

menu-services-egochores.jpg is on the first screen but marked loading="lazy", so the browser waits before fetching it and the page looks finished later. Load it eagerly (and give it fetchpriority="high" if it is the main image).

6 images further down the page (670 KiB) download before anyone scrolls to them. Adding loading="lazy" lets the first screen load first.

4 images are downloaded at several times the size a phone draws them, for example face-01.jpg at 1,024 px wide, shown at 58 px. Resized copies would save about 350 KiB.

By the Sustainable Web Design model, a visit to your homepage emits about 1.14 g of CO₂e: 137.0 kg a year at 10,000 visits a month. It scales with page weight, so every byte cut lowers it.

2 tap targets are too small or too close together to hit reliably with a finger.

All page content should be contained by landmarks: 24 elements on the page fail this check, rated moderate by axe-core.

The page has no <main> landmark, so screen-reader users cannot jump straight to the content.

6 visible links have no text or label, so screen readers cannot say where they go.

No canonical URL, so duplicate versions of the homepage (with or without www, with tracking parameters) can split ranking signals.

The homepage links to 68 pages on heartex.ai and 11 external pages.

/blog/vibe-code-any-labeling-or-evaluation-interface-with-label-studio-enterprise, /platform, /blog/introducing-human-in-the-loop-evaluation-for-agentic-ai-observability, /ai-benchmarks and /blog/improving-rag-document-search-quality-with-cohere-re-ranking use the same meta description. A description written for each page earns more clicks.

We fetched 15 pages linked from your homepage (of 41 found) and read each the way a search engine does.

2 of 8 tappable elements near the top of the page are smaller than 24 px, hard to hit with a thumb.

6 errors were logged to the browser console while the page loaded, for example “Failed to load resource: the server responded with a status of 404 ()”.

No Content-Security-Policy header, the main browser-side defence against injected scripts.

X-Content-Type-Options: nosniff is missing, so browsers may guess file types.

Opening https://heartex.ai goes through 2 redirects before the homepage loads.

2 links go through two or more redirects, for example https://docs.humansignal.com/.

No Referrer-Policy header, so full URLs can leak to the sites you link to.

No Permissions-Policy header to switch off camera, microphone and location for embedded content.

There is no /.well-known/security.txt telling researchers how to report a security problem to you.

humansignal.com has no CAA record, so any certificate authority may issue a certificate for it. A CAA record naming your provider (for example letsencrypt.org) closes that door.

humansignal.com's DNS answers are not signed. DNSSEC stops attackers forging them to send visitors or email elsewhere; most registrars switch it on in a click, though most sites still run without it.

On a phone, there is no call to action on the first screen: visitors have to scroll to find out what to do next.

Social proof on the homepage: 6 customer or partner logos. Pairing it with named testimonials or customer logos makes it far more convincing.

No phone number or email link on the homepage, so ready-to-buy visitors have one fewer way to reach you.

There is no form on the homepage, so every lead has to click through to another page first.

No guarantee or risk reducer (money-back, free trial, no credit card) is mentioned, which is often what tips a hesitant visitor.

The homepage carries about 296 words of visible text.

Email authentication: SPF record 25 points, plus 15 when it ends in -all (10 for ~all); DMARC record 25, plus 25 for a reject policy (15 for quarantine); a DKIM key on a common selector 10.

Before a visitor clicks anything, the page loads Google Ads, Google Analytics, Google Tag Manager, HubSpot and LinkedIn Insight and 4 mores (measured from a US location before any click). A consent tool (HubSpot cookie banner) is installed; check it blocks these until visitors agree where the law requires it.

Opening the homepage sets 27 cookies (measured from a US location before any click).

Loading the homepage contacts 45 hosts run by other companies (fonts, scripts, analytics, ads), each of which sees the visitor's IP address.