Design and diagrams decision
Makebanner
A competent engineer can reproduce the essential product (prompt UI, style mapping, and calls to a hosted image API) in about a week and maintain it cheaply; no durable moats are shown on the site.
Visit website↗$89/mo
$1,068/yr
Read off the official pricing page.
$100one-off40 h to build
$10/mo4 h/mo upkeep
On cash alone, building overtakes the subscription at 1 seat.
No open-source build does this yet
Nothing published replaces this one, so a replacement starts from an empty file. Here is what it would have to cover.
What a replacement has to do
- User enters one-sentence prompt → agent picks matching art styles → call image-generation API to produce 3 variations in selected aspect ratios → format and provide download
What it still won’t have
- Proprietary style-matching refinements and any curated style library
- Polish of the vendor UI (priority queues, batch generation flows)
- Dedicated support and SLA
- Any proprietary model weights if used by the vendor
What remains hard
- Product polish and ongoing maintenance
First-year cost
Keep paying
Paying is—cheaper in year one.
On cash alone, building overtakes the subscription at 1 seat.
Money you would actually spend
Time you would spend
—
What you would spend
What we assumed
The verdict above measures whether you could build it. This one is only about money.
Runnable build prompt
Build an AI banner generator as a Next.js app with a small Node/Express API and Postgres for storage. Core features in scope: single-line prompt UI, aspect-ratio selector, a style-catalog and a deterministic style-matching function, integration with a hosted image-generation API (e.g., Stable Diffusion/Replicate) to produce 3 variations per request, image post-processing to export high-res PNGs in 10 aspect ratios, a simple credit-based payment flow using Stripe (one seat), signed URLs for downloads, and an admin dashboard to view usage. Out of scope: training new generative models, enterprise multi-tenant billing, and advanced collaboration features. Include input validation, retries/backoff for API calls, error handling, basic unit/integration tests, and deploy scripts for Vercel (frontend) and a small server on Render or Vercel Serverless, with documented environment variables and monitoring setup.
How we checked
How the score was reached
- Build verdict base78
- 2 cited sources+1
- Price verified on pricing page+3
- Evidence score82
The base comes from the verdict. Everything under it is a check that either happened or did not, and each one is a fact frozen in this record rather than a judgement made at render time - so the same evidence always produces the same number.
How scoring works →Cited sources · 2
Every page the run actually retrieved.
- official productmake-banner — official product page
- official pricingmake-banner — pricing
Integrity checks
What held up, and what did not.