Public audit / not a customer

Discover Conversion Leaks from Real SaaS Landing Page Audits

These companies are not Nebula customers. We run the same 9-signal engine on public pages and publish the raw findings - not to criticize anyone, but to show exactly what the audit produces on pages you can verify yourself.

Public audit report

carrd.co

Carrd

4.2/10

Grade D

5 findings

H1 is <h1 className="heading-1 tracking-tight text-fg md:text-5xl">Discover Conversion Leaks from Real SaaS Landing Page Audits</h1> - the brand name only. Zero value prop. Missing og:title, og:description, and all JSON-LD schema.

Audited July 30, 2026

Read teardown →

Public audit report

hotjar.com

Hotjar

4.5/10

Grade D

4 findings

H1 announces an acquisition: "Hotjar has evolved into something more powerful." First-time visitors must decode what Hotjar does before seeing why to convert.

Audited July 30, 2026

Read teardown →

Public audit report

kit.com

Kit

4.8/10

Grade D

4 findings

Meta description is 35 chars. Industry average is 140-155. Google auto-generates the SERP snippet - Kit has ceded their first impression in search entirely.

Audited July 30, 2026

Read teardown →

Public audit report

beehiiv.com

Beehiiv

5.1/10

Grade C

5 findings

4 competing CTAs above fold before the hero. H1 is a product list with no verb, no outcome. 262KB homepage HTML with canonical mismatch.

Audited July 30, 2026

Read teardown →

Public audit report

unbounce.com

Unbounce

5.2/10

Grade C

4 findings

H1: "Launch faster. Convert more." - identical positioning to Leadpages, Instapage, and Swipe Pages. The company that invented the landing page builder has a commodity headline.

Audited July 30, 2026

Read teardown →

Public audit report

webflow.com

Webflow

5.8/10

Grade C

4 findings

Two H1 tags simultaneously in the DOM via A/B test. Google indexes both variants - ranking signal is split between two competing headlines.

Audited July 30, 2026

Read teardown →

Evidence-backed teardown

framer.com

Framer

7.8/10

Grade B

4 findings

Primary CTA reads "Start without AI" on an AI-first product page. Leads with an opt-out of the core differentiator. H1 missing from static HTML - JS-render dependency.

Audited July 30, 2026

Read teardown →

Public audit report

basecamp.com

Basecamp

5.4/10

Grade C

5 findings

<title> is 8 chars. Meta desc truncates at 155. No named proof above fold.

Audited July 29, 2026

Read teardown →

Public audit report

notion.so

Notion

6.1/10

Grade C

4 findings

417KB HTML payload (3.4× heuristic max). H1/title keyword misalignment.

Audited July 29, 2026

Read teardown →

Public audit report

calendly.com

Calendly

5.8/10

Grade C

5 findings

Missing H1 entirely. 1.2MB HTML payload - 10× the heuristic ceiling.

Audited July 29, 2026

Read teardown →

Public audit report

postmint.de

Postmint

6.8/10

Grade B

2 findings

Founder published a full launch autopsy (118 visitors, 0 signups) - both self-diagnosed bugs are fixed. One real defect remains: the JSON-LD @context key is a leaked Blade/PHP template literal, not "@context".

Audited July 31, 2026

Read teardown →

Public audit report

knallhart.dev

knallhart.dev

5.8/10

Grade C

3 findings

Zero social-sharing metadata (0 og: tags, 0 twitter: cards) on a product whose founder's entire distribution is X posts and forum threads - every shared link renders as a bare URL.

Audited July 31, 2026

Read teardown →

Public audit report

hubspot.com

HubSpot

7/10

Grade B

5 findings

H1 and title share zero keywords - and the title ends in "- Homepage" (a template placeholder). The static H1 search engines index ("grow scale close retain grow") differs from the headline human visitors see.

Audited August 4, 2026

Read teardown →

Public audit report

mailchimp.com

Mailchimp

7.2/10

Grade B

5 findings

Six H1s on one page, each telling a different story about what the product is. Plus a first-visit personalization modal that covers the hero before a new visitor reads a single word.

Audited August 4, 2026

Read teardown →

Public audit report

cloudcostwise.io

CloudWise

7.1/10

Grade B

3 findings

Founder-reported: approximately $2,700 in ad spend, around 6 trial signups, and 0 paid conversions. Current snapshot: strong page fundamentals, but ad-tracking evidence is not visible in static source.

Audited August 3, 2026

Read teardown →

Public audit report

postdew.com

PostDew

4.8/10

Grade D

5 findings

Founder-reported: 0 paying customers, 0 signups from approximately 10 LinkedIn cold DMs, and 0 signups from a flagged Show HN post. Current snapshot: no static H1, CTA, or trust markers.

Audited August 3, 2026

Read teardown →

Public audit report

smartwatermark.app

SmartWatermark

7.5/10

Grade B

2 findings

Founder-reported: 84 landing-page visitors, 24 app visitors, 0 sales, and 79% bounce. Current snapshot: strong headline, CTA, SEO, and proof signals; runtime tracking remains unverified.

Audited August 3, 2026

Read teardown →

Public audit report

www.folioverse.app

Folioverse

5.6/10

Grade C

5 findings

Founder-reported: €100 in Google Ads spend, 91 clicks, 0 signups, and 10-second average paid-visitor sessions. Current snapshot: no static H1 or CTA candidate, with ad tracking unverified.

Audited August 3, 2026

Read teardown →

Public audit report

asana.com

Asana

6.9/10

Grade B

4 findings

Strong above-fold with clear CTA but social proof is generic, no named customer outcomes above the fold.

Audited August 2026

Read teardown →

Public audit report

canva.com

Canva

5.7/10

Grade C

5 findings

Headline is feature-first, not outcome-first. No conversion tracking artifact found in static HTML.

Audited August 2026

Read teardown →

Public audit report

figma.com

Figma

7/10

Grade B

4 findings

Above-fold is strong but meta description truncates at 160 chars, topic relevance signal is cut off.

Audited August 2026

Read teardown →

Public audit report

intercom.com

Intercom

6.9/10

Grade B

4 findings

CTA is action-oriented but positioned below a long feature list, most mobile visitors never reach it.

Audited August 2026

Read teardown →

Public audit report

monday.com

Monday.com

6.9/10

Grade B

5 findings

Social proof is present but proof markers are vague counts, not specific named outcomes.

Audited August 2026

Read teardown →

Public audit report

shopify.com

Shopify

7.1/10

Grade B

4 findings

Strong SEO foundations and load speed but headline is brand-focused, not buyer-outcome focused.

Audited August 2026

Read teardown →

Public audit report

slack.com

Slack

6.6/10

Grade B

5 findings

Message match is weak on paid traffic, the headline does not reflect typical ad copy promises.

Audited August 2026

Read teardown →

Public audit report

squarespace.com

Squarespace

7.2/10

Grade B

4 findings

Above-fold is clean but no analytics tracking artifact detected, paid traffic attribution is unverified.

Audited August 2026

Read teardown →

Public audit report

typeform.com

Typeform

6.2/10

Grade C

5 findings

Headline communicates the product category, not the buyer outcome. No price or effort signal above fold.

Audited August 2026

Read teardown →

Public audit report

zapier.com

Zapier

7.4/10

Grade B

3 findings

Strong overall but Open Graph tags incomplete, social sharing previews are degraded.

Audited August 2026

Read teardown →

Evidence-backed teardown

linear.app

Linear

7.3/10

Grade B

3 findings

H1 source contains two line-break collapses: "productdevelopmentsystem" and "teamsand" fused into single words, same defect class as Shopify, Asana, and Squarespace. Invisible in the browser, indexed by search engines as garbled text.

Audited August 2026

Read teardown →

Public audit report

loom.com

Loom

7.2/10

Grade B

3 findings

Title tag, "Free screen recorder for Mac and PC | Loom", positions Loom as a screen recorder while the page pitches team communication and async video. Category mismatch between the SERP promise and the page's actual positioning.

Audited August 2026

Read teardown →

Evidence-backed teardown

miro.com

Miro

7.3/10

Grade B

3 findings

H1 is "The collaboration layer your AI tools are missing.", a positioning claim framed as absence. "Missing" implies Miro is invisible by default, an unusual negative framing for a primary conversion headline.

Audited August 2026

Read teardown →

Public audit report

clickup.com

ClickUp

5.8/10

Grade C

4 findings

H1 is "Software to replace all software.", no product category, no audience signal, no outcome. The claim is so broad it carries zero conversion signal for anyone searching for a specific tool.

Audited August 2026

Read teardown →

Evidence-backed teardown

airtable.com

Airtable

7.3/10

Grade B

3 findings

Title appends "| Airtable" after already opening with "Airtable:", brand name appears twice in 65 characters. Title reads "Airtable: Build Enterprise-ready AI Workflows, Apps & Agents | Airtable". 65 chars also truncates in most SERPs at ~60.

Audited August 2026

Read teardown →

Public audit report

pipedrive.com

Pipedrive

7.4/10

Grade B

3 findings

One of the cleaner above-folds in this batch: H1, proof claim ("Trusted by over 100,000 companies"), CTA, and friction-removal copy all in the hero. Primary weakness: email capture field in the hero duplicates the button CTA with no visual hierarchy differentiation.

Audited August 2026

Read teardown →

Public audit report

mixpanel.com

Mixpanel

6.4/10

Grade C

4 findings

H1 is "Build faster, with direction", no product category, no audience named. Above-fold also carries the tagline "Mixpanel is for teams that move. Make your move.", a second vague statement that also fails to name what the product does.

Audited August 2026

Read teardown →

Public audit report

amplitude.com

Amplitude

6.2/10

Grade C

4 findings

Title is "Amplitude | A new era for product teams", 38 characters, generic "new era" framing, no product category keyword. The hero opens with a 17-logo trust strip before the value proposition, asking cold visitors to anchor to customer names before understanding the product.

Audited August 2026

Read teardown →

Public audit report

drift.com

Drift

5.4/10

Grade C

4 findings

drift.com now serves a Salesloft/1mind product page with the notice: "We've transitioned from Drift to 1mind." Visitors searching for "Drift" land on a different product's pitch with no redirect explanation or legacy-brand acknowledgment above the fold.

Audited August 2026

Read teardown →

Evidence-backed teardown

deel.com

Deel

8.9/10

Grade A

0 findings

Deel’s homepage covers global hiring, payroll, HR, and equipment with a strong category promise, but its unusually large document and broad product surface deserve a more explicit first path for different buyers.

Evidence-backed teardown

raycast.com

Raycast

6.8/10

Grade B

0 findings

Raycast explains the breadth of its shortcut and command model, but its first-screen message and action path require visitors to do more interpretation than the product needs.

See what it finds on your page

Free, no signup. Same engine as every teardown above.

Find the Leak →