In our sample, Lovable sites score higher: a median of 61/100 against 49/100 for Bolt, a gap of 12 points. The score counts problems a visitor or a search engine would hit, such as missing headings, exposed keys and blocked zoom, so it says as much about how people use each builder as about the builder itself. One side is a small sample, so treat the gap as a hint rather than a verdict.
Side by side
from our study of 780 public sites
Lovable
Bolt
Type
AI builder (site written from a prompt)
AI builder (site written from a prompt)
Default hosting
*.lovable.app
*.bolt.host
Sites in our sample
200
18
Median site score
61/100
49/100
Sites with a critical problem of their own
38%
33.3%
Call a Supabase database from the browser
34.5%
33.3%
No main heading (H1)
70.5%
77.8%
No meta description
0.5%
5.6%
Zoom disabled on phones
2%
0%
Builder badge left on
66.5%
0%
Highlighted: the better of the two on that row. Lower is better for every percentage.
Telling them apart
Lovable
How to spot it: Open the page source (Ctrl+U) and search for gpteng.co. Lovable projects load their editor script from that host.
What usually goes wrong: API keys left in the JavaScript bundle. Lovable projects often talk to Supabase or another backend straight from the browser, so the key ends up in a file anyone can download. If that key is a service role key rather than an anonymous one, the whole database is readable. Our scanner downloads up to three bundles and reports any key it finds, redacted.
How to spot it: Look at the address: *.bolt.host is Bolt hosting.
What usually goes wrong: Supabase keys in plain sight. We found a Supabase key in the JavaScript bundle of most Bolt sites we scanned. An anonymous key is meant to be public, but only if row-level security is switched on in the database. Many projects never turn it on, and then the key is enough to read every row.
Can you tell whether a site was built with Lovable or Bolt?
Usually, yes. Each leaves its own traces: Lovable on *.lovable.app and in its scripts, Bolt on *.bolt.host and in its own markers. Paste the URL in the scanner above and it names the builder with the evidence it found.
Which is better for SEO, Lovable or Bolt?
Neither decides it alone. In our sample the biggest SEO gaps were a missing main heading and a missing meta description, both fixable on either platform. Compare the rows above, then scan your own site to see which of them apply to it.
Is a Lovable site worse than a hand-built one?
Not by being AI-built. The problems we see most are defaults nobody changed: titles, badges, blocked zoom, and database keys used without access rules. Each is a short fix once you know it is there.
Numbers from The AI-built web, October 2026: public sites on each builder's own hosting, scanned once. Read the method and its limits there.