Build · Create · Analysis & Finance
Normalized competitor comparison scorecard
The buyer gets a structured comparison grid with a distinct row for every competitor they name, matched on whole words against the row's name or domain, so one merged row cannot stand in for two competitors. Each row carries the competitor's name, a host-like domain, site traffic, a G2 rating, an authority rank, a positioning phrase, an ease-of-use note, and one specific weakness. The row count must reach at least the number of named competitors, domains are unique across rows, the numeric metrics are checked as in-range, and the written entries must clear minimum word floors rather than merely being present.
You receive: A JSON array of competitor objects, each with name, domain, site_traffic, g2_rating, authority_rank, positioning, ease_of_use, and specific_weakness.
Part of Launch Product
Example
A sample of what this play produces. Your result is generated for your inputs.
| Name | Domain | Site traffic | G2 rating | Authority rank | Positioning | Ease of use | Specific weakness |
|---|---|---|---|---|---|---|---|
| ChatGPT | chatgpt.com | 3800000000 | 4.7 | 94 | The default general-purpose AI chat assistant: a blank prompt box for any task, sold on raw model capability rather than any packaged outcome. | Trivial to start, hard to master: results depend on prompt craft, and the buyer re-explains their business in every session. | Nothing compounds between sessions: output lands as loose chat text the founder must re-check and carry somewhere by hand, with no acceptance check screening what counts as done. |
| Fiverr | fiverr.com | 58000000 | 4.5 | 92 | A freelance gig marketplace selling fixed-scope human deliverables, competing on price and seller ratings across a long-tail catalog. | Browsing gigs is easy, but scoping, briefing each seller from scratch, and managing revisions puts real coordination work on the buyer. | Every gig starts from zero context: a founder rewrites the same business brief for each seller, turnaround is measured in days, and quality varies seller to seller with no structured output that later work can reuse. |
| Upwork | upwork.com | 40000000 | 4.6 | 91 | An hourly and contract freelance marketplace aimed at longer engagements, positioned on vetted professional talent for teams. | Posting a job is simple, but interviewing, contracting, and supervising freelancers is a hiring workflow most solo founders do not have time to run. | Cost and latency scale with hours, not outcomes: a small structured deliverable can cost hundreds of dollars and a week of back and forth, versus a credit-priced play that lands in one run. |
| Notion | notion.so | 130000000 | 4.7 | 90 | A flexible docs and workspace tool where teams store what they know, increasingly bundling AI features on top of static pages. | Easy to write in, but the buyer designs every structure themselves: templates are empty scaffolds the founder still has to fill and maintain. | Documents hold knowledge but do no work: pages go stale the day after they are written, and its AI features summarize what is already there rather than producing new checked deliverables that feed later work. |
Opens soon
STUD opens runs for everyone soon, on every play at once.
Scope of the check
What's verified: STUD verifies the scorecard parses as a JSON array with at least as many rows as the competitors you name (or your minimum row count, default 1, whichever is higher), that each competitor you name is matched, on whole words, in its own distinct row's name or domain (one merged row cannot cover two names, a short name like 'Box' is not covered by 'dropbox.com', and a name that contains no letters or digits cannot be matched and fails the check), that every row carries all eight fields, that each domain is host-like (no scheme, no spaces) and unique across rows, that site_traffic is a non-negative integer, g2_rating is a number from 0 to 5, and authority_rank is an integer from 0 to 100, and that the written entries are real text: the name at least 2 characters, positioning and ease_of_use at least 3 words each, and the specific weakness at least 4 words. STUD does NOT verify the traffic, rating, or authority figures against real-world data, does NOT judge whether the positioning or weakness statements are true or insightful for your market, and does NOT check that your product context or target customer was used; those remain the operator's craft.
Get early access to STUD the day it goes live.