/labs/scruuge-ai
← Back to /labs
IDEA FORGE LABS · RUBRIC-FLIP STUDY

Scruuge.AI — AI Cost Optimization

Security and ops burden don't shrink at lifestyle scale.

Confidence delta: 20 pts
THE IDEA · INPUT TO BOTH AUDITS

Scruuge.AI — an AI cost optimization SaaS that sits between users' applications and their AI providers. Routes API calls through OpenRouter to the cheapest model meeting a quality threshold. Charges 20% of verified savings only — if no savings, no charge. OpenAI-compatible drop-in proxy with Zero Data Retention. Two routing tracks: same-quality-lower-cost (Track A, default) and cheapest-possible (Track B, opt-in). Target: indie developers spending $50-500/month and small businesses spending $200-2K/month on LLM API costs.

Two rubrics. Two verdicts.

Venture rubric
Large TAM. Moats. Scaling unit economics.
BORDERLINE
51%
The savings-share model is genuinely clever, but the trust premise it requires — "you can't verify what we did, just send us money" — is exactly the wedge OpenRouter and the model providers themselves are closing in real time.
Smart pricing model in a category where the value gap is actively shrinking.
The 20%-of-savings model is the strongest part of this idea. It mirrors Cloudability and Vantage in cloud-cost optimization (real businesses, $100M+ ARR each), and it solves the trust problem at signup ("if no savings, no charge") that makes most B2B AI tools hard to convert. But three structural forces are closing the savings gap before Scruuge can capture it. (1) OpenRouter's own routing has been improving aggressively; "smart router" is now their default. (2) Anthropic and OpenAI have cut prices 60-80% over 2024-2025, with Haiku 4.5 at $1/$5 per Mtok and gpt-4o-mini at $0.15/$0.60. The sav…
Read full venture rubric report →
Lifestyle rubric
$1–5k/month. Under 10 hours/week.
CONCERNS
31%
*This idea's lifestyle rubric score is worse than its venture score — not because the rubric is lenient, but because the specific structural problems (security-review ops, linear-to-revenue maintenance, competition from venture-funded incumbents) do not shrink when you shrink the ambition.*
Lifestyle scale doesn't reduce the ops burden — it removes the leverage to survive it.
The lifestyle rubric asks one structural question: can one person reach $1–5k per month within six months while working fewer than 10 hours per week? For Scruuge as pitched, the answer is no — but the reason is worth stating precisely because it differs from the venture objection. The venture rubric's concern is the closing savings ceiling. The lifestyle rubric's concern is ops-that-scales-with-revenue: a proxy architecture means every customer who onboards brings a security conversation, a potential outage incident, and an ongoing debugging surface. At 10 customers those conversations take 15…
Read full lifestyle rubric report →
HOW IS THIS HONEST? · WHY THE SAME IDEA GETS TWO VERDICTS

The audit didn't change its mind. It answered a different question.

Venture rubric asks

Could this be a fundable, scaling business?

Lifestyle rubric asks

Could one person at <10h/week reach $1–5k/month?

Same facts. Inverted signal weights. The audit doesn't reconsider the evidence — it reweights it. What counts as a positive signal under one rubric can be a fatal negative under the other:

  • Small TAMconcern (no path to scale)fine (only ~100 customers needed at $20/mo)
  • No moatconcern (incumbents will copy)fine (organic discovery + niche knowledge IS the moat)
  • 6–12 month sales cycleacceptable for B2B SaaSfatal (no revenue within time budget)
  • Ops linear to revenuefixable with team at scalefatal (no time budget for support)
  • Security-review burdenamortize over many customersfatal (same friction at any scale)

Why this matters for honesty. A single-rubric service that defaults to venture framing would tell a stay-at-home parent or a side-hustler that their idea has “no moat” or “small TAM” — technically correct, but irrelevant to their actual goal. That's being right inside the wrong question. We'd rather ask the question first.

Read the methodology → · See all 8 ideas →

Run a Reality Check on your idea · $29 →We'll ask your goal profile up front and evaluate against it specifically.