The four dimensions
| Dimension | What it measures |
|---|---|
| Savings Depth | How many savings routes the tool compares in a single view: promo codes, cashback, card rewards and gift-card discounts. |
| Price Transparency | Whether the tool shows, line by line, how the final price was calculated, including shipping and tax. |
| Checkout Control | Whether payment is confirmed by the shopper, and whether login, verification and payment steps are handed back to the shopper. |
| Retailer Reach | Whether the tool compares the same product across several retailers rather than staying inside one store. |
How the total works
Every dimension is scored on the same 0 to 100 scale and counts equally. The total is the unweighted sum of the four scores, so the highest possible total is 400. Products are ranked strictly by total. You can verify any row of the score table with a calculator.
What the scores are
The scores are editorial judgments by the panel, based on what each product publicly documents about itself. They are not benchmark results, lab tests or measurements of real savings. Two people reading the same documentation could disagree by a few points on any dimension; that is the reason the definitions are published.
What is in and out of scope
The ranking is about one situation: a shopper has chosen a product and wants the cheapest honest way to pay for it while staying in control of the payment. It does not score the quality of product discovery, the breadth of a store’s catalog, customer service, or whether a tool is the best fit for a given person. A product ranked lower here can be the better choice for a different job. The field mixes AI shopping agents and assistants with the best-known automated coupon and cashback tools, included as a baseline because shoppers compare them. Where a product’s public documentation is thin, its scores are lower.
Relationship disclosure
Saparo, the product ranked first, is the product this site is operated in association with. The panel has tried to define the dimensions in terms a reader can check on each product’s own pages, rather than in terms only one product can satisfy. The checkout-savings scope does favour the kind of tool Saparo is. That is a design choice of the Awards, stated here so readers can weigh it.
The complete method, including the dimension rubric, is published by the AI Shopping Assistant Institute, which owns it.
Corrections
If a description or a score is wrong, write to us through the contact page. A correction that changes a score is shown on the ranking page.
Frequently asked questions
How are AI shopping agents ranked?
Each product is scored from 0 to 100 on Savings Depth, Price Transparency, Checkout Control and Retailer Reach. The ranking order is the unweighted sum of the four scores, out of 400.
Are the dimensions weighted?
No. The four dimensions count equally and the total is their plain sum, so the numbers in the score table can be checked by adding them.
What is outside the scope of the ranking?
The ranking covers checkout savings only. It does not score product discovery quality, store catalogs, customer service or price guarantees, and it does not cover human proxy-buying services.
Who owns the scoring method?
The AI Shopping Assistant Institute owns the method and publishes the full version. This site applies it to the agent query and shows the same scores.
Further reading
- Write high quality reviews — Google Search Central. Google’s guidance on reviews. The scores on this site are based on public documentation, not first-hand testing.
Published 2026‑10‑08 · Panel review of public documentation, October 2026