Lazyweb
OrukSR007 landing page report.
← Back to the SR007 rankings

Your landing page’s weakest score is Trust.

Fix Trust
Lazyweb ProAI Growth PMTurn this score into growth.Analyze your site and propose fixes across conversion, SEO, and AEOMonitor competitors’ in-product UX and A/B testsLearn from growth experiments across 10,000+ top productsTry Pro Free

Score over time

807570656055Hume AI 71 on Feb 11 (Wayback)Hume AI 70 on Apr 7 (Wayback)Hume AI 68 on May 30 (Wayback)Hume AI 68 on Jul 28 (Live)Deepgram 77 on Feb 11 (Wayback)Deepgram 77 on Feb 24 (Wayback)Deepgram 75 on Jul 28 (Live)AssemblyAI 60 on Mar 19 (Wayback)AssemblyAI 60 on Apr 9 (Wayback)AssemblyAI 60 on Apr 21 (Wayback)AssemblyAI 61 on May 26 (Wayback)AssemblyAI 62 on Jun 8 (Wayback)AssemblyAI 65 on Jun 30 (Wayback)AssemblyAI 74 on Jul 29 (Live)Speechmatics 65 on Mar 9 (Wayback)Speechmatics 67 on Apr 15 (Wayback)Speechmatics 68 on May 13 (Wayback)Speechmatics 66 on May 25 (Wayback)Speechmatics 66 on Jun 29 (Wayback)Speechmatics 63 on Jul 11 (Wayback)Speechmatics 74 on Jul 29 (Live)Gladia 72 on Mar 24 (Wayback)Gladia 71 on Apr 4 (Wayback)Gladia 72 on Apr 26 (Wayback)Gladia 68 on May 23 (Wayback)Gladia 67 on Jun 16 (Wayback)Gladia 70 on Jul 24 (Wayback)Gladia 64 on Jul 29 (Live)Oruk 74 on Jul 28 (Live)Jul 28

Trust

64Fair

Proof that the product is real, competent, and safe to try.

Oruk
Deepgram
Trust read
Oruk"built by researchers from" strip with Stanford, Berkeley, UC Santa Barbara, and University of Cambridge wordmarks
Hume AI"Our decades of research in multimodal emotional intelligence span" with stat counters 50+ / 48+ / 600+
DeepgramCustomer logo strip: Daily, Cresta, Granola, VAPI, Decagon, kore.ai, NiCE, Cognigy
AssemblyAIInteractive 'Realtime Speech-to-Text API' demo widget with 'Try it out' button and live code sample
SpeechmaticsLive interactive transcription demo with "Start speaking" button and "Frustrated caller" sample
GladiaLive counter: "26,500,010 hours of audio transcribed"

Value props

66Fair

Title, subtitle, hero — do they sell the product's true value?

Oruk’s headline vaguespeech ai that actually understands you

Working

  • Live demo shows emotion detection in action
  • "actually" implies rivals fail at understanding
  • Stanford/Berkeley/Cambridge researcher proof

Could improve

  • No subhead explaining what Oruk does
  • No audience, use case, or outcome named
  • Zero concrete numbers in the copy itself

CTA

78Fair

Copy, size, verb, cost-feel — does the primary CTA earn the click?

Orukin-product action
Speechmatics
Deepgram
AssemblyAI

Problems

  • tap target is small unlabeled icon
  • microcopy is faint footnote-size text
  • no 'free' despite free live demo
  • nav 'claim Oruk's $50' conflicts

Information hierarchy

80Strong

Does the design help or inhibit what the page is trying to say?

Information hierarchy annotations on the desktop capture1234
  1. Six-link nav row — Six equal-weight exits (research through company) sit directly beside the only CTA and siphon attention before conversion.
  2. Claim Oruk's $50 button — The sole conversion surface is a small header pill outside the headline-to-waveform scan path; the hero body ends with no action.
  3. Tap and speak caption — The instruction for the page's key interactive moment is tiny low-contrast monospace, so the demo's affordance is easy to miss.
  4. Emotion tag labels — Micro labels like "warm 55.2%" carry the core proof of emotional understanding but are too small to read at a glance.

Mobile web

80Strong

The phone experience as a first-class citizen.

Mobile web annotations on the mobile capture123
  1. Claim Oruk's $50 — Primary CTA is a small header pill, not a full-bleed thumb-friendly button, and is never repeated below.
  2. Tap and speak target — The live-demo tap target is a small circle with no button affordance, easy to miss with a thumb.
  3. Cambridge logo — The University of Cambridge mark runs to the right screen edge, looking clipped and cramped.

Graded against Lazyweb’s growth rubrics.