candy.ai website review

71 images could be 3.6 MB smaller with modern formats (WebP or AVIF) and sizes that fit a phone screen. Google estimates it would make the page about 4.0 s faster on a phone.

About 793 KiB of JavaScript is downloaded but never used on this page. Google estimates it would make the page about 3.9 s faster on a phone.

Your mobile performance score is 45. Under 50 is Google's “poor” band: many phone visitors leave before the page is usable.

21 files are served with short cache lifetimes, so returning visitors download them again (806 KiB).

2 files block the first paint until they finish downloading. Deferring or inlining them could save about 800 ms.

6 long tasks block the main thread while loading, the longest for 302 ms. During a long task the page ignores taps.

The page has 2,608 HTML elements. Above 1,500, rendering and every interaction get slower on phones.

A phone's processor spends 2.4 s on main-thread work while loading, 990 ms of it running JavaScript. Above 2 s, taps and scrolls feel sluggish.

6 third-party services load with your homepage (Google Tag Manager, Cookie-Script.com and Other Google APIs/SDKs and others), adding 442 KiB and 210 ms of main-thread time.

Polyfills for old browsers add 25 KiB JavaScript that modern phones do not need. Google estimates it would make the page about 150 ms faster on a phone.

About 12 KiB of CSS is downloaded but never applied on this page.

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 17.3 s to render on a phone. Google's threshold for “good” is 2.5 s. The largest element is the image.

The first text or image appears after 6.8 s on a phone. Good is under 1.8 s.

The page becomes fully interactive after 17.6 s. Good is under 3.8 s.

Visually, the page fills in at 9.6 s (Speed Index). Good is under 3.4 s.

Real visitors wait 2.5 s for the largest content to appear (75th percentile). Good is under 2.5 s.

Real visitors wait 460 ms for the page to respond to a tap (75th percentile). Good is under 200 ms.

Real visitors see the first content after 2.3 s (75th percentile). Good is under 1.8 s.

Real visitors wait 1.3 s for your server to start answering (75th percentile). Good is under 800 ms.

Scripts block taps and scrolling for 439 ms in total while the page loads. Good is under 200 ms.

Over the last 25 four-week periods to 2026-10-03, real Chrome visitors' tap response (inp) went from 315 ms to 382 ms at the 75th percentile (Chrome UX Report (opens in a new tab) history).

Over the last 25 four-week periods to 2026-10-03, real Chrome visitors' largest content (lcp) went from 2.9 s to 3.3 s at the 75th percentile (Chrome UX Report (opens in a new tab) history).

Your homepage downloads 5.3 MB in 176 requests on a phone: images 3.9 MB (148), scripts 1.1 MB (8) and stylesheets 163 KiB (3). Under 2 MB keeps phones on slow connections fast.

21 images are downloaded at several times the size a phone draws them, for example neon-run-logo-44ddecc60bec264c297de8e69fb0e5168c52478a77eb4a at 777 px wide, shown at 76 px. Resized copies would save about 1.1 MB.

By the Sustainable Web Design model, a visit to your homepage emits about 1.51 g of CO₂e: 181.5 kg a year at 10,000 visits a month. It scales with page weight, so every byte cut lowers it.

Certain ARIA roles must contain particular children: 1 element on the page fails this check, rated critical by axe-core.

1 button has no accessible name, so screen readers announce just “button”.

20 text elements don't have enough contrast against the background to read easily, especially outdoors on a phone.

<dt> and <dd> elements must be contained by a <dl>: 13 elements on the page fail this check, rated serious by axe-core.

Some list items are not inside a list.

34 images have no alt text, which hurts accessibility and image search.

All page content should be contained by landmarks: 3 elements on the page fail this check, rated moderate by axe-core.

The viewport tag blocks pinch-zoom, which low-vision visitors rely on.

Alternative text of images should not be repeated as text: 11 elements on the page fail this check, rated minor by axe-core.

1 form field has no label, so screen readers announce an empty box and autofill guesses.

Elements with visible text labels do not have matching accessible names. (53 elements).

Your meta description is 228 characters; Google truncates after about 160, so the end of your pitch is cut.

1 link has no real href, so search engines can't follow it.

1 link use generic text like “click here” or “learn more”, which tells Google nothing about the page behind it.

The homepage links to 51 pages on candy.ai and 15 external pages.

There is no https://candy.ai/llms.txt, a short Markdown guide to your key pages for AI tools (llmstxt.org). It is optional and few AI tools read it yet, so treat it as a cheap extra.

Your homepage links to https://candy.ai/characters/new2, which answers with an error. Visitors hit a dead end and search engines drop it.

/, https://candy.ai/subscriptions, https://candy.ai/ai-girlfriend/lea-fontaine/live-actions and https://candy.ai/ai-girlfriend/nari-ryo/live-actions all use the title “Candy AI: AI Girlfriend App | Chat w/ Pics, Voice, & Video”. Search engines can't tell which page to show for a search, so each competes with the others.

https://candy.ai/community-guidelines has no meta description, so Google writes the search snippet from whatever text it finds.

https://candy.ai/subscriptions, https://candy.ai/ai-girlfriend/lea-fontaine/live-actions and https://candy.ai/ai-girlfriend/nari-ryo/live-actions use the same meta description. A description written for each page earns more clicks.

Your homepage links to 2 pages by a URL that isn't their canonical one, for example https://candy.ai/ai-girlfriend/lea-fontaine/live-actions, whose canonical is https://candy.ai/ai-girlfriend/lea-fontaine. Linking straight to the canonical URL saves crawlers a detour and keeps ranking signals on one address.

https://candy.ai/feed/posts, https://candy.ai/subscriptions, https://candy.ai/ai-girlfriend/lea-fontaine/live-actions and https://candy.ai/ai-girlfriend/nari-ryo/live-actions have no <h1> headline, so visitors and search engines get no clear main topic.

https://candy.ai/feed/posts and https://candy.ai/characters have under 150 words in the HTML. Thin pages rarely rank; if the text loads with JavaScript, crawlers that don't run it see even less.

We fetched 15 pages linked from your homepage (of 17 found) and read each the way a search engine does.

2 errors were logged to the browser console while the page loaded, for example “Failed to load resource: net::ERR_FAILED”.

Large scripts ship without source maps.

Your homepage shows a street address but no LocalBusiness JSON-LD, so Google and AI assistants have to guess your address, hours and phone from the page text.

A local customer looking to call finds no phone number on the homepage. Add one, ideally as a tap-to-call link.

Social proof on the homepage: 4 trust badges. 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.

The homepage shows no opening hours. If customers visit or call, show when.

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

There is no link to Google or Apple Maps. A directions link is one tap for a phone visitor.

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 1,821 words of visible text.

Before a visitor clicks anything, the page loads Google Tag Manager and Mixpanel (measured from a US location before any click). No consent tool was detected. Visitors from the EU, UK and several US states must be asked first.

Opening the homepage sets 4 cookies (measured from a US location before any click).

No known cookie consent tool loads with the page. Some sites build their own banner, and some only show one to visitors from the EU.

Loading the homepage contacts 7 hosts run by other companies (fonts, scripts, analytics, ads), each of which sees the visitor's IP address.