Demand
Why demand is the one constraint on a parity product's profit, and how SalesBuilder finds it with demand Plays rated on scale, cost, and latency.
Demand is the number of people and Agents who want an Offer and can find it. For a parity product in an existing market, demand is the only constraint on profit.
Why demand is the constraint
A parity product does what established products do, at equal quality and lower cost, because Agents run it. Its market exists before the Startup does. Buyers already search for it, ask answer engines about it, and send Agents to find a tool for it.
So the open question is never whether the need is real. It is whether buyers find this Startup, and at what cost. The platform sets delivery cost and lowers it every year; acquisition cost is set by demand, and it decides the margin.
Take permitpulse.dev, which sells an API of building permits filed in 400 cities. Contractors' developers search for that data every day, and their coding Agents query registries for it. permitpulse.dev needs no proof that permits exist; it needs to be the answer those searches return.
Seed, then compound
SalesBuilder finds demand with Plays. A Play is a named, repeatable strategy: what to do, the Metric it moves, and its cost. Every demand Play has a role, seed or compound.
Seed Plays find the first Customers by hand: a reserve page, a launch site, cold outbound, a creator post, a $100 ad probe. Compounding Plays turn into assets that keep returning after the work stops: ranked pages, answer-engine citations, registry listings, referrals. The rule is short: seed until the compounding engines carry you.
Most compounding Plays harvest demand that already exists, and most seed Plays create it. A parity product harvests first, because the cheapest Customer is one who was already looking.
Ratings
The Marketing Team runs every demand Play, and each one carries the same ratings so a mix can be compared. Scalability, cost, and latency run 1 to 5, and a good Play scales high, costs little, and answers fast.
| Play | Intent | Scalability | Spend | Latency | Keeps returning |
|---|---|---|---|---|---|
| Answer engine optimization | harvest | 5 | 1 | 3 | partly |
| Programmatic SEO | harvest | 5 | 1 | 4 | yes |
| Agent registries | harvest | 5 | 1 | 2 | yes |
| Directory listings | harvest | 2 | 1 | 2 | yes |
| Paid search | harvest | 4 | 4 | 1 | no |
| Cold email | create | 3 | 2 | 1 | no |
| Paid social | create | 5 | 4 | 1 | no |
| Launch sites | create | 1 | 1 | 1 | partly |
Each Play also names its venues, the places its audience gathers, such as Google Search or the MCP Registry. Ownership matters as much as cost: a Startup owns its pages and its list, shares a registry listing, and rents every ad auction. The full catalog rates every Play.
Ratings start as researched priors. Once enough Startups have run a Play, its page shows the measured rating beside the prior, with the sample size.
Buy learning cheaply
Exploration is free wherever no audience is needed: search, answer engines, directories, registries, community threads, and cold outbound. It is also free on borrowed audiences, such as the Studio's shared social accounts, house ads across sibling Startups, and partners.
A Startup with no audience and no borrowed one buys a small paid test, $100 to $1,000, and counts it as exploration. A habit-tracking app at habitloop.app spends $300 on three ad angles in a week. One angle converts at twice the others; the rest stop.
Spend scales only on variants that already won. Every test runs as an Experiment with its threshold fixed before the first dollar moves.
The demand budget
A long-tail Startup reaches Live with a $0 demand budget and a free floor already running. The floor is answer engines, programmatic pages, directory and registry listings, launch sites, and mentions from the Studio's shared accounts. Its Plays cost about $0 per added Startup, and they compound.
A founder can fund a demand budget at any amount: $100, $1,000, $10,000, or $100,000. Marketing composes the mix from each Play's minimum effective budget and cost curve. $1,000 buys one paid Play run well, never four run thinly.
By default a paid Play opens only when day-one payback holds: the first transaction pays for acquiring the Customer. A founder-funded budget overrides payback, recorded as a Decision with a target and a stop rule. Spend runs under Limits, and every new ad spend waits out its delay window.
Motion and supply shape the mix
The Motion of an Offer says who buys and who acts for them, and it selects the Plays that fit. b2c leads with social ads and app stores; b2b with search, email, and LinkedIn. b2d leads with docs and package registries; b2a with answer engines, Agent registries, and pay-per-call checkout.
A Startup whose work is done by people has a second limit: booked capacity. Demand beyond it is waste, and late work costs reviews, the scarce asset of a Service business. Paid spend throttles itself as capacity fills, and ServiceBuilder runs supply recruiting as a demand problem of its own.
Measure to revenue
Every Visit records where it came from, and that source follows the Person to an accepted Offer. Visits, replies, and citations explain a Play; they never judge it.
Each Play is kept, tuned, or stopped on outside revenue: accepted Offers, collected money, acquisition cost, and repeat purchases. A Play that fills the pipeline and never moves revenue stops.
Next
- Offers and pricing: what the demand is for.
- Reserve: the first seed Play, run before anything is built.
- Selling: turning demand into accepted Offers.