Skip to content
SiteList

Supabase Review: strong platform, clear gaps (75/100)

Supabase scores 75/100, with clear product positioning and substantial documentation for developers building applications. Its most material weakness is mobile performance, where sampled p75 LCP is 5.94 s and INP is 598 ms.

Reviewed by SiteList Engine · 12 of 13 dimensions · published Reviewed on October 25, 2026

Is this your site?

Claim it

Quick facts

Supabase is an open-source Postgres development platform with integrated backend services and vector embeddings.

Field Value
Domain supabase.com/modules/vector
Category Open-source Postgres development platform with integrated backend services and vector embeddings
Pricing Free; Pro from $25/month; Team from $599/month; Enterprise custom pricing
Pages crawled 38
Crawl date 2026-10-18
Evidence
Overall SiteList Review Score
75/100
Pages crawled
38
Crawl date
2026-10-18

Executive summary

Supabase presents a clear open-source Postgres development platform for developers and product teams building production applications. First impressions scored 81.2/100, usability 87.8/100, accessibility 86/100, and docs and self-serve help 87.4/100. The supplied crawl also found a verified sitemap containing 3,665 URLs and a technical SEO score of 91/100.

The main drag is uneven execution rather than unclear positioning. Design execution scored 46/100. Performance scored 60.8/100: sampled content showed p75 LCP of 5.94 s and INP of 598 ms, while CLS was 0.04 and TTFB was 533 ms. Writing quality scored 67.6/100, decision-support surfaces 68/100, and editorial QA 64.2/100. Those results point to a platform that is easy to classify and research, but whose mobile experience and supporting editorial layer need more consistency.

Evidence
First impressions & positioning
81.2/100
Docs & self-serve help
87.4/100
Technical SEO
91/100; sitemap 3,665 URLs
Mobile performance signal
p75 LCP 5.94 s; INP 598 ms; CLS 0.04; TTFB 533 ms

01 · First impressions & positioning — “Build in a weekend” names the outcome, not the buyer

Supabase is easy to classify as an open-source Postgres development platform, but its hero leaves the primary buyer implied. “Start your project with a Postgres database” establishes the category, while “Build in a weekend Scale to millions” gives the promise a memorable speed-and-scale frame. The product grid then makes the integrated mechanism concrete with Authentication, Edge Functions, Storage, Realtime, Vector embeddings, and REST APIs. Proof exists—“10,000,000+ Registered developers” and “100,000+ GitHub stars”—but it sits on the company page rather than beside the homepage claim. Add one audience-specific clause for developers or product teams and move one attributable proof point into the hero area. Preserve the visible comparison path for Auth0, Heroku Postgres, and Firebase; it gives the positioning useful competitive context.

Evidence
Homepage hero
Build in a weekend Scale to millions
Registered developers
10,000,000+
GitHub stars
100,000+

02 · Audience & messaging — Free, Pro, Team, and Enterprise are clear; fit remains broad

Supabase gives a developer enough information to identify the offer, find pricing, and start a project quickly, but audience fit and operational trade-offs are not separated clearly. The homepage names Postgres, APIs, Auth, Functions, Storage, Realtime, and Vector, while pricing provides Free, Pro from $25/month, Team from $599/month, and Enterprise with custom pricing. The broad promise could serve a weekend builder, startup team, or larger engineering organization, yet those paths are mostly implied. Comparison pages for Auth0, Heroku Postgres, and Firebase supply useful rationale, while “Use one or all. Best of breed products. Integrated as a platform.” leaves the trade-off compressed. Add audience paths and a compact evaluation FAQ covering migration, limits, cancellation, support, and production recovery.

Evidence
Pricing entry points
Free; Pro from $25/month; Team from $599/month; Enterprise custom pricing
Homepage product set
Postgres, APIs, Auth, Functions, Storage, Realtime, and Vector

03 · Usability — 8.8 s mobile LCP on Vector interrupts an otherwise direct path

Supabase offers a strong first-time path: “Build in a weekend / Scale to millions,” “Start your project,” Pricing, and Docs give a developer clear next moves. Pricing exposes $0, $25, $599, and Custom options, while documentation provides “Connect a framework” and “Search docs...” cues. The main friction is measurable on the Vector page: the mobile PSI artifact reports 8,777 ms LCP, 12,704 ms TTI, 5,839 ms FCP, and 913 ms TBT. General contact is also footer-dependent: the homepage foregrounds “Request a demo,” while Contact Us appears in the footer graph. Improve the Vector first interaction, add Contact or Support to the primary path, and replace repeated pricing “Get Started” labels with plan-specific actions.

Evidence
Vector mobile LCP
8,777 ms
Vector mobile TTI
12,704 ms
General contact discovery
Contact Us appears in the footer link graph

04 · Accessibility — strong sampled structure, with docs landmarks and contrast still failing

Supabase’s sampled pages have several verified template issues that need correction. Lighthouse reports missing html lang and main landmarks on the JavaScript and Swift reference URLs, plus a color-contrast failure on Vector. Heading sequences also skip levels, including h2→h6 on Contact, h1→h3 on Pricing, and h1→h3 plus h3→h2 on Vector. Repair the docs reference template, correct heading nesting, add the docs skip link, and remeasure touch targets and contrast. Live keyboard and screen-reader checks remain pending.

Evidence
Docs reference Lighthouse failures
html-has-lang and landmark-one-main on two sampled URLs
Vector color contrast
PSI color-contrast score 0
Sampled missing-alt images
0

05 · Design execution — 89 small tap targets make the mobile system feel mechanical

Supabase’s captured layouts establish a polished baseline, but interaction sizing and structural consistency hold back the visual system. The desktop and mobile Vector runs fail color-contrast audits, and the mobile facts record 89 small tap targets, a 40×40px menu button, a 26px skip link, a 34px input, and 12px input text. The same sample reports no horizontal overflow at 390px, so the problem is not basic containment. The homepage outline also moves from H4 to H6 in the footer, which PSI flags as a heading-order failure. Raise audited hit areas to at least 44×44px, make the input at least 44px high with 16px text, and correct semantic heading levels. Treat the inventory of 305 colors and 153 font sizes as an audit lead until computed styles confirm rendered drift.

Evidence
Small mobile tap targets
89
Mobile input
34px high; 12px computed font size
Vector mobile layout
scrollWidth=390 at a 390px viewport

06 · Performance — p75 LCP 5.94 s and INP 598 ms make mobile the limiting case

Supabase needs a faster mobile critical path: sampled docs and content pages report p75 LCP of 5.94 s and INP of 598 ms, while CLS is 0.04 and TTFB is 533 ms. The lab traces point to client-side and blocking work rather than layout instability. The sampled blog reaches 7.2 s LCP, 793.5 ms TBT, and 2.76 s of main-thread work, with an estimated 620 KiB of unused JavaScript and 1.34 s of render-blocking savings. Payloads are also large: Vector is 6.1 MB across 76 requests and the homepage is 11.5 MB across 136. Prioritize critical content, split route JavaScript, defer non-essential hydration, and set per-template budgets. Desktop Vector field results—1.66 s LCP and 68 ms INP—show the gap is concentrated in mobile.

Evidence
Sampled mobile p75 LCP
5.94 s
Sampled mobile INP
598 ms
Vector network summary
6.1 MB across 76 requests

07 · Writing quality — specific developer language is weakened by repeated structure and long sentences

Supabase writes most effectively when it names a developer action: the Vector page says “Use pgvector to store, index, and access embeddings” and names Hugging Face, SageMaker, and OpenAI. Pricing similarly gives concrete limits such as 50,000 monthly active users and 500 MB database size. The editorial drag is consistency at scale. Vector, Cron, and Queues move from H1 directly to H3 feature headings, while extracted metrics report average sentence lengths of 73 words on Pricing and 103 on Blog. The crawler also reports the homepage description shared by four pages and the pricing description shared by two. Keep the specific technical voice, but correct heading levels, split long commercial copy, give metadata page-specific wording, and replace generic phrases such as “Easily connect” with a concrete workflow or constraint.

Evidence
Pricing average sentence length
73 words
Blog average sentence length
103 words
Shared homepage description
4 pages

08 · Decision-support surfaces — 49-row pricing matrix needs a concise choice layer

Supabase provides unusually concrete decision material, but the path from evidence to recommendation is too deep. Pricing exposes Free, Pro, Team, and Enterprise with quotas, overage rates, compute sizes, connection limits, storage economics, and support or compliance differences. The detail then expands into a 13-row compute table, a 49-row comparison table, and a larger feature matrix. The Pro “Most Popular” label does not explain when a buyer should override it. Keep the technical tables for expert readers, but lead with scenario-based guidance: Free for experiments, Pro for production teams needing predictable quotas, Team for SSO, compliance, and support, and Enterprise for negotiated controls. Add last-reviewed dates, comparison axes, and source notes; Auth0 is dated 2023-11-24 and Heroku Postgres 2022-05-26.

Evidence
Pricing comparison table
49 rows
Compute table
13 rows
Auth0 comparison date
2023-11-24

09 · Risk & stability — 3,665 sitemap URLs and long history support a stable public surface

The supplied evidence shows a stable, crawlable public surface, with mobile documentation performance as the clearest loss-capable risk. Robots.txt allows crawling and declares the sitemap, sampled public pages return 200, a nonexistent URL returns 404, and the two noindexed paths are account surfaces: /dashboard and /dashboard/sign-up. The sitemap contains 3,665 URLs, and Wayback coverage runs from 2018 through 2026 without supplied evidence of an unhealed migration or domain repurposing. The risk to monitor is narrower: two reference templates report mobile field LCP of 5,943 ms and INP of 598 ms, with 980px horizontal overflow at a 390px viewport. Fix that shared template, then compare mobile clicks and impressions in Search Console before and after release. Traffic magnitude cannot be measured from this crawl.

Evidence
Verified sitemap
3,665 URLs
Wayback coverage
2018 through 2026
Reference-template mobile LCP
5,943 ms

10 · Editorial QA of content — technical specificity is strong, but template checks are uneven

Supabase clears the basic editorial bar with useful technical specificity, real authors and dates in supplied BlogPosting schema, and executable SQL on the Cron and Queues pages. The QA layer is not normalized across templates. The company page reports 40 H1 elements, module pages repeat H1-to-H3 skips, and the crawler finds missing descriptions on dashboard surfaces alongside duplicate titles and descriptions elsewhere. The module family also repeats a recognizable sequence—Postgres Native features, a “100% Open Source” claim, then SQL, API, and dashboard sections—which helps orientation but reduces distinctiveness when every page sounds alike. Add a shared gate for one H1, valid H2/H3 nesting, metadata uniqueness, alt coverage, and provenance for high-salience statistics. Require each module opening to include one scenario, boundary, or failure mode.

Evidence
Company heading count
40 H1 elements
Module template pattern
Postgres Native, 100% Open Source, SQL/API/dashboard sections
Community claim
10,000,000+ Registered developers

11 · Docs & self-serve help — current support is strong, but two SDK references are orphaned

Supabase treats documentation as a first-class product surface, with clear paths for getting started, product areas, references, migrations, self-hosting, AI, support, and changelog updates. The changelog entry dated Oct 14, 2026—four days before the crawl—shows active maintenance, and support offers Discord, GitHub issue or discussion paths, and ticket escalation. The edge problem is findability. The JavaScript and Swift reference URLs have inDegree=0 despite an organized Reference category, so users may need direct URLs to reach them. The docs root also needs a behavior check: rendered facts report no search box or breadcrumbs while the screenshot shows “Search docs...”. Add contextual links to both references, verify the search control in the accessibility tree and keyboard flow, and preserve the 3,242-byte llms.txt index as documentation expands.

Evidence
Latest changelog entry
Oct 14, 2026
Orphaned reference pages
2 sampled URLs with inDegree=0
llms.txt body
3,242 bytes; status 200

12 · Technical SEO — 3,665 URLs are crawlable, but mobile reference templates overflow

Supabase has a solid technical SEO baseline: robots.txt allows crawling, declares the sitemap, and the verified sitemap contains 3,665 URLs. HTTP probes consolidate the HTTP and www variants to the canonical HTTPS host, sampled HTTPS responses carry HSTS, and raw versus rendered text is nearly identical on the homepage at 1,358 versus 1,360 words and exactly identical on Vector at 895 versus 895. The main technical defect is mobile documentation rendering. Two reference URLs lack a viewport declaration, show 980px scroll width at a 390px viewport, and report field LCP of 5,943 ms and INP of 598 ms. Changelog and Firebase comparison samples also overflow at 390px. Add responsive viewport coverage, constrain wide content, and re-run the full template family after the fix.

Evidence
Verified sitemap
3,665 URLs
Homepage raw/rendered text
1,358 / 1,360 words
Vector raw/rendered text
895 / 895 words

Verdict — 75/100: strong platform, clear gaps

Supabase earns 75/100 because its product position is clear, its public usability is strong, and its documentation gives developers substantial self-serve coverage. The site explains the core offer, supports evaluation with visible plan information and comparisons, and presents a technically strong crawlable surface.

The fixable weaknesses are concentrated in execution: design execution scored 46/100; sampled mobile performance showed p75 LCP of 5.94 s and INP of 598 ms; and editorial QA scored 64.2/100. Supabase is a good fit for developers and product teams that want an open-source Postgres platform with integrated backend services and vector embeddings. Teams making a decision should give extra weight to the mobile research experience and the depth of the supporting editorial layer.

Evidence
Overall score
75/100
Design execution
46/100
Editorial QA of content
64.2/100

Methodology & data notes

This 13-dimension review uses the supplied crawl of 38 pages, product facts, public dimension scores, one-line dimension summaries, and the listed evidence excerpts. The crawl date is 2026-10-18. Scores are reported from the supplied public score table and are not re-scored in this article frame.

Dimension 13, Review-content integrity, was excluded as not applicable. Google Search Console was not connected and access was unavailable, so search-console enrichment remains pending. The review does not claim evidence that was not present in the supplied inputs. Read How SiteList scores for the scoring method and data notes.

Evidence
Review scope
13 dimensions
Crawl sample
38 pages crawled on 2026-10-18
Excluded dimension
13 · Review-content integrity — not_applicable
Enrichment status
GSC connected: false; GSC access: false

Questions buyers actually ask

What is Supabase?

Supabase is an open-source Postgres development platform with integrated backend services and vector embeddings. The supplied product facts identify web and API platforms, plus Authentication, Data APIs, Edge Functions, Realtime Data, Storage, and Vector embeddings.

Who is Supabase for?

Supabase is aimed at software developers and product teams building web or mobile applications, from weekend projects to internet-scale production.

How strong is Supabase's documentation?

The documentation score is 87.4/100. The crawl found organized Start, product, build, manage, reference, resources, self-hosting, AI guides, support, and comparison material.

Is Supabase easy to evaluate?

The crawl found visible plan prices, included quotas, overage rates, compute sizes, and competitor comparisons across meaningful technical axes.

What is Supabase's main weakness in this review?

Mobile performance is the clearest weakness: sampled content showed p75 LCP of 5.94 s and INP of 598 ms. Design execution also scored weak at 46/100, so mobile presentation and performance are the first areas to improve.

How this review was made

SiteList reviewed supabase.com on October 25, 2026 — pages, screenshots, performance runs, structured data and public records — then scored it across 12 of the 13 public dimensions. Every claim above is sourced from what we collected; nothing is hand-tuned and the score is never for sale.

Not covered in this write-up: Review-content integrity (not applicable). Dimensions without a score are excluded and their weight is redistributed across the scored ones.

Pending enrichment (data we could not fetch this run): Brand SERP samples were unavailable because no SERP endpoint was configured., Live click execution and form-validation behavior were not available., Mobile screenshots were available for the homepage, but not for every task destination., Live axe-core verification on representative templates, Scripted keyboard tab-order and focus-trap testing, Screen-reader announcement, caption accuracy, and dynamic error-state testing, Raw computed contrast-pair inventory with foreground, background, ratio, and selector, Complete per-element tap-target and spacing inventory across all sampled templates…

Read the full methodology

75/100SupabaseJump to review