Skeleton + animated placeholder
Candidate skeleton while OpenAI call runs; first result 1.2s, all results 3.5s.
AI Utility / Generative Tool · Case Study
Wheelie Names — AI Brand & Game Name Generator
An OpenAI-API-powered name generator for brands, products, and game concepts — fast discovery via micro-animations + share-sheet integration + AEO/SGE-aligned snippet structure.
Results
Measurable Outcomes
1.2s
First name render time
3.5s
All 20 names
Domain + 2× social
Per-name validation
TODO — analytics
Share rate
The Brief
What was the goal?
The name-generator market is saturated; most competitors do basic word mixing. In the AI era, "creative + applicable" names need an extra layer: domain availability + entity disambiguation + tone.
The Approach
Which methodology pillars I applied
I applied the AEO ("how to come up with a game name" FAQ) + Schema (SoftwareApplication + AggregateRating) + GEO (semantic chunking, citation-ready format) pillars.
See the full process → 47-Point AI-Ready SEO Audit methodology.
Tech Stack
What it's built with
- Next.js 15
- TypeScript
- OpenAI API
- Whois API
- Schema.org SoftwareApplication
The Build
Key decisions and tradeoffs
While the OpenAI API call is running, instead of leaving the user on a blank screen, we render a 12-candidate skeleton with animated placeholders. First name appears at 1.2s; all 20 land at 3.5s — drop-off rate is low.
Next to every generated name: a domain availability badge (Whois API) + Twitter/Instagram handle availability. This eliminated the "creative but unusable" problem.
Share-sheet integration: when a user likes a name, one tap shares to X/Threads/LinkedIn + a customized Open Graph card. That created a user-generated traffic loop.
Steps Applied
Methodology in practice
Domain + handle availability
Whois API + social handle check per name; eliminates "creative but unusable" friction.
Share sheet + custom OG
One-tap X/Threads/LinkedIn share + custom Open Graph card; user-generated traffic loop.
Honest Reflection
What I'd Do Differently
On v1 we didn't block new form submissions while the OpenAI call was in flight. Impatient users smashed submit 2-3 times, eating API cost every time. Monthly OpenAI bill came in 4× higher than projected. On v2 I disabled the submit button during in-flight requests, added debounce + cache. Cost-per-user dropped from $0.18 to $0.04.
Screenshots
Want this for your project?
Same 47-point methodology, $499 fixed price, 5-7 business days. Book the audit.