One system, nine stages, one loop.
Each stage answers a question the previous one raised, and hands the answer forward as structured data — not as a document someone re-keys into the next tool. The last stage feeds the first, which is why the ranking you see next month is different from the one you see today.
- 01Audience
- 02Intelligence
- 03Opportunity
- 04Concept
- 05Economics
- 06Sales
- 07Operations
- 08Experience
- 09Analytics
Who is actually out there?
Every platform you are on holds a fragment of the same person. Audonomics pulls in social, video, music, email, CRM, commerce and ticketing data, resolves the handles back to one record, and keeps an append-only log of what that person actually did.
- Identity resolution across platforms — one member, many handles, one history
- An append-only event log: the substrate every score and opportunity is built on
- Location, first seen, last seen, lifetime spend, orders, events attended, referrals
- Import by connector or by file; every source is labelled and re-syncable
Who among them is worth building for?
Rules, not vibes. Members fall into ten segments and carry a 0–100 value score produced by a model whose weights you control and whose arithmetic you can read line by line.
- Casual, Engaged, Customer, Repeat, VIP, Superfan, Community Leader, Traveler, Collector, Experience Buyer
- Configurable value scoring — edit a weight, see every score move and why
- A visible breakdown on every member: which inputs contributed how many points
- Predicted lifetime value is stored separately and labelled Modeled, never Observed
What is the highest-value thing we could create?
The engine reads the audience continuously and ranks what you could build for it — a destination experience, a retreat, a cruise, a festival, a private dinner series, a membership, a drop, a mastermind, a tournament, a multi-city run.
- Scored on revenue potential, margin, demand, audience fit, complexity, risk and time-to-launch
- Every score opens into its factors and weights — there is no black box
- Each opportunity carries a thesis and machine-checkable reasons, each tagged with provenance
- Shortlist, dismiss or promote; dismissals feed back into the next ranking
What exactly is it?
A promoted opportunity becomes a concrete thing with a shape: type, dates, locations, legs, venue, capacity, and the tiers you intend to sell.
- Experience blueprints across gather, immersion, access, participation, travel, commerce, membership and education
- Multi-leg by default — a tour, a cohort across cities, a series of pop-ups
- Tiers with price, unit cost, quantity, benefits, deposit terms and sales windows
- Capacity weighting, so one suite can consume four seats of the room
Does the math actually work?
A real contribution-margin model, not a revenue guess. Fixed costs, marginal cost per unit, tier mix, break-even, minimum viable capacity and the conversion rate you would need to hit it.
- Conservative, Base and Upside cases computed from the same inputs
- Break-even and minimum viable capacity stated in guests, not percentages
- A scenario simulator: move sell-through or a tier price and the model recalculates instantly
- Money is integer minor units end to end — no float ever touches the money path
Will they actually pay?
Validate demand before you commit capital: waitlists, refundable deposits, presales and tier tests. Real signals feed straight back into the model. Then open inventory that cannot oversell.
- Waitlist, interest, deposit, presale, survey and vote campaigns — each tied to a tier
- Signals update tier mix and sell-through assumptions in the financial model
- Inventory integrity enforced by a database CHECK constraint, not by application code
- Deposits, installment plans, upgrades, transfers, refunds, discounts and referral codes
Can we execute it?
An approved opportunity becomes an executable project: milestones, tasks, dependencies, budget lines, vendors, contracts, staffing and a run of show — with the schedule reasoning about itself.
- A dependency engine that answers: if the date moves two days, what breaks?
- Budget lines reconciled against real vendor quotes and contracts
- RFQs to multiple vendors, quote comparison, contract approval trail
- Risks registered against the project, with owners and mitigation
Does it land on the day?
Attendees, travel, lodging blocks, transport segments, communications, community and the minute-by-minute run of show — all reading from the same records that sold the thing.
- Lodging and transport assignment that cannot be over-allocated
- Segmented communications and notifications from live attendee data
- Community spaces and groups scoped to an experience or a membership
- Run of show with owners, cues and times
What did we learn?
A post-event report built from actual order and cost records, plus survey responses — and then the loop closes: what happened updates who your audience is, and the ranking changes.
- Realised P&L against the model that justified the decision
- Attendance, tier mix and refund behaviour written back onto member records
- Survey responses attached to the experience and the attendee
- The next ranking is produced by an audience that now includes this event
What a financial model looks like when it is real.
Three cases from one input set. Change the marginal cost per guest and all three move together, because they are outputs of the same arithmetic rather than three separate spreadsheets someone kept in sync by hand.
- Fixed costsVenue hold, production, staff travel, insurance
- $48,200
- Marginal cost per guestLodging night rate, meals, ground transport
- $780
- Blended priceWeighted across three tiers at planned mix
- $2,993
- Contribution per guestBlended price less marginal cost
- $2,213
- Break-evenFixed costs ÷ contribution per guest
- 22 guests
- Minimum viable capacityBreak-even plus contingency reserve
- 26 guests
| Case | Conservative | Base | Upside |
|---|---|---|---|
| Sell-through | 45% | 72% | 95% |
| Guests | 27 | 43 | 57 |
| Gross | $80,800 | $128,700 | $170,600 |
| Variable costs | $21,060 | $33,540 | $44,460 |
| Fixed costs | $48,200 | $48,200 | $48,200 |
| Contribution | $11,500 | $47,000 | $77,900 |
| Margin | 14% | 37% | 46% |
Example product surface. Figures are illustrative.
The things the system will not do.
A platform that handles money, contracts and other people’s travel plans should be defined as much by its refusals as by its features.
It will not present an estimate as a fact
Every derived number carries a provenance label, and confidence is capped by the weakest input feeding it.
It will not let an agent spend your money
Anything creating a financial, contractual or legal commitment sits in an approval queue until a human with the right role approves it.
It will not oversell your inventory
Sold plus held can never exceed quantity. That is a database constraint, not a check in application code that a race condition can slip past.
It will not round your money badly
Money is integer minor units everywhere. Installment schedules always re-sum to the order total; refunds can never exceed what was paid.
Why the last stage is the first stage.
The post-event report is not an archive. Attendance, tier mix, refund behaviour and survey responses are written back onto member records, which changes segment membership, which changes value scores, which changes the ranking.
- 160 members attend the Lisbon retreat. 43 had never travelled for you before.
- 2Their records gain events attended, lifetime spend, and travelled-for-event = true.
- 343 members move from Engaged into Traveler; 11 cross the Superfan threshold.
- 4The Traveler segment grows from 511 to 554, with a higher median value score.
- 5Tomorrow's ranking puts a second destination above the festival that led last month.
Illustrative. The mechanism is real; the numbers are examples.
Start at stage one.
Connect a source and let the engine tell you what your audience is actually worth building for.