Aiso field model · ROI
Which AI-search actions should you fund first?
First published .
The break-even formula is simple
Monthly contribution from AI = AI-referred sessions × conversion rate × contribution margin per conversion.
For one intervention, estimate the incremental contribution it could create, then divide implementation cost by that monthly increment. That gives a break-even period. The hard part is not the arithmetic. It is refusing to call correlation causation.
Particle is the clean public outcome we can point to: its AI visibility increased approximately fivefold during the Aiso engagement, as reported by mind Retail. The engagement combined on-site work, schema, crawlability, content, and off-site source work. That result does not tell us which individual tactic caused what share of the lift. That distinction matters.
Interactive worksheet
Estimate break-even, not vanity ROI
Change the inputs. The lift is an assumption until you measure it.
This is a planning model, not an Aiso performance guarantee. Attribute the observed lift only after the intervention has enough post-change data.
Fund the first broken link, not the trendiest tactic
| Action | What you measure first | When it deserves budget | Time to proof |
|---|---|---|---|
| Retrieval fixes | Pages become fetchable; schema validates; bots reach the right URLs | Do first when the evidence is technically unreachable | Fast |
| Third-party source work | Relevant independent pages begin mentioning the brand or correcting facts | Do early when assistants lean on outside sources in the category | Medium |
| Answer-shaped content | New pages cover the real decision questions and fan-outs currently missing | Do when the model can retrieve you but has no page that answers the buyer's constraint | Medium |
| Listing consistency | Names, descriptions, URLs and attributes agree across direct and third-party surfaces | Do early for travel, local, ecommerce and marketplace-heavy categories | Fast to medium |
A worked example
Suppose AI assistants currently send 1,000 sessions a month. They convert at 2.5%, and each conversion contributes $120 after variable costs. That traffic is worth $3,000 a month in contribution.
If one intervention plausibly adds 20% more qualified AI traffic, the modeled increment is $600 a month. A $1,800 implementation breaks even in three months. Those are illustrative inputs, not an Aiso benchmark.
The same math works for a conversion-rate intervention. If the number of AI visits stays flat but better landing-page alignment moves conversion from 2.5% to 3.0%, model that lift instead. Do not add a traffic lift and a conversion lift unless you have a reason to believe both happen independently.
What I would measure before revenue arrives
Revenue is a lagging metric. For each action I want an earlier proof signal: successful bot retrieval, a new cited source, a new brand mention on the target prompt set, a shift in which competitor is recommended, or a measurable change in AI-referred sessions.
This is why a cheap technical fix can beat a large content program. If the right page already exists but cannot be reliably fetched, writing ten more pages increases the pile rather than fixing the bottleneck.
Download the worksheet
Download the AI-search ROI template as CSV. It includes the five calculator inputs plus columns for the intervention, pre-change date, post-change date, evidence signal and attribution notes.