Frequently asked questions

A pre-release product from RateHighway that is designed to reveal and measure how aware AI agents are of a brand and location pairing, with respect to car rental booking and availability questions. The unit is one brand at one airport. The output is a coverage map, not a score.

No. This is a pre-release. Every figure in the application is simulated so reviewers can react to the experience. The official version is planned for 2027.

Your name, a work email, your organization, the brands you operate from the fixed catalog, and up to three airport markets on the Free plan. Sign-in runs through AWS Cognito only. The Free plan also needs your own AI provider key before any observation runs.

An airport code is unambiguous, it is how rate shopping is already organized, and it is how a traveler frames the question. City and regional markets are planned later.

Type the three-letter airport code. We validate it against the IATA list and show it as "LAX - Los Angeles International" from then on.

Free: your own AI key, three markets, fifteen prompts, weekly observation, local visibility and position KPIs. Standard: $25 per month including two markets and $12 per month for each additional market, 25 prompts, weekly, evidence and exports. Pro: 35 prompts, twice weekly, root cause and retest. Pro+: Pro plus a concurrent rate shop so quoted prices are compared with live rates.

[STUB: important question, answer pending.] Owner: product and finance (OQ-32).

In advance, each month, on the anniversary of your first paid charge. If that day is the 29th, 30th or 31st we bill on the 28th so every month has the day.

Yes. Upgrades and added markets take effect immediately and you pay a prorated amount for the rest of the current month. Downgrades and removed markets take effect on your next anniversary. You always see the exact preview, including the next bill date and amount, before anything is charged, and that preview is kept with the change.

We retry and tell you before observation stops. Nothing is deleted. Your history stays readable.

[STUB: important question, answer pending.] Owner: finance (FR-SU-11).

Yes, from the billing page. Access continues to the end of the paid period and you can export your findings throughout.

Observation means asking an AI assistant questions on a schedule, and each question costs a little. With your key, you pay your provider directly for those questions and RateHighway supplies the prompts, schedule, parsing and KPIs for free.

1. Create an account with the AI provider you want to test. 2. Open the developer or API section of your account. 3. Add a payment method there. 4. Set a monthly spending limit. 5. Create a new key named "Agent Awareness" and copy it once. 6. Paste it into Agent Awareness. We check it with one small request. 7. You can revoke it any time, here or at the provider.

[STUB: important question, answer pending.] Provider names appear only once RateHighway is approved to display them. Owner: legal (OQ-31).

Our internal estimate for a weekly run of fifteen prompts across three markets, repeated three times each with live web search, is roughly $7 to $40 a month depending on the model you choose, with the web search fee the largest part. [STUB: important question, answer pending.] The published range will come from measured usage.

Encrypted with a RateHighway-held key. We keep only the last four characters for recognition, never write it to a log or a URL, and decrypt it only inside the run that uses it.

Ten fixed traveler questions, one per intent class: availability and booking asked by name, open availability, cheapest price, total price, terminal counter versus shuttle, vehicle, under-25 policy, local brands and one-way. Eight do not name your brand, because the finding is whether the agent volunteers you.

It turns your brands and markets into realistic traveler prompts for a chosen trip purpose and traveler type, from RateHighway-written templates. You pick the ones to keep. Standard and above can save them into custom prompt slots.

Write the way a traveler types: dates, airport, what matters to them, no jargon. Ask the same need several ways. Keep brand-named and unnamed versions. Do not stuff the prompt with your brand, because the point is to learn what the agent says on its own.

Yes. Every edit creates a new version and results stay attached to the version that produced them, so a trend is never silently broken.

The share of unnamed observations at a market in which the agent mentioned your brand for that market. Why it matters: the agent volunteered you without being asked.

The 0 to 100 expression of Eligible Awareness Rate: recognized eligible observations divided by eligible observations, where eligible means the pairing was verified sellable. Why it matters: it separates "the agent did not know we were there" from "we had nothing to sell". It needs availability truth and is locked until that exists.

Where your brand sits in the answer when mentioned, and how often it came first. Why it matters: in a short list, position decides whether the traveler reads you.

The share of observations that cited a given OTA, broker, direct site or other source. Why it matters: it shows who the agent trusts for your market.

Whether the price the agent quoted matches your current rate within a freshness window, and whether the quoted total is within tolerance of the real total. Both need RME truth, through Pro+ or a connected RME account.

Local is your own markets. National is every observed airport in your country. Global is every observed airport everywhere. National and global views always state how many airports were observed and how many were not.

[STUB: important question, answer pending.] Owner: revenue management (OQ-33).

Travelers increasingly put the whole question to an assistant: where to rent, what it costs, whether a car is available, and how to book. The assistant compresses a long comparison into a short answer. An operator who is not in that answer is invisible, even with the fleet, the rate and a working booking path.

Its picture of your airport comes from what it has read and from the sources it trusts: OTAs, brokers, your own site, review and mapping platforms. If those sources are thin, stale or inconsistent about you at that location, the agent has nothing to go on.

No. Answers vary by phrasing, day and personalization. Only repeated observation across many prompts separates a real gap from noise, which is why the product samples repeatedly and never presents a single test as representative.

[STUB: important question, answer pending.] Owner: product research.

Bring your airports under observation, read the local view for a few weeks, then look at which sources the agent cites for your market. Distribution and listing accuracy are the levers you already control.

Inconsistent location facts across sources, missing or stale OTA listings, an off-airport lot described as if it were on airport, and policy details that contradict each other. Each cause carries a confidence label when the product proposes it.

Pro and above retest the same prompts under comparable conditions and show before and after. A changed answer after a fix is evidence, not proof, and is labeled that way.

[STUB: important question, answer pending.] Owner: product, once real observations exist.

In the region you choose at signup: the United States, or the European Union in Frankfurt. Personal data, keys, prompts and evidence stay in that region. Only anonymous brand and market aggregates are shared to build the national and global views.

Everything is stored in UTC. Your browser decides how dates and times are displayed, and the UTC value is always available on hover.

The time in UTC, your IP address, an approximate location derived from it, your browser and its language and time zone, and whether the sign-in succeeded. You can see your last twenty sign-ins under Settings, Security.

Owners can export findings at any time and request deletion. Deletion completes within thirty days.

Yes. The owner can terminate the account from Settings, Account. The subscription is canceled first, your data stays exportable for thirty days, and then it is deleted. Anonymous brand and market aggregates that never carried your account remain.

Yes. A person owns exactly one account and can be a view-only guest on many. Each account has a friendly name its owner chooses, and the account switcher lists them by that name.

Yes. Upload your brand’s logo and pick its colors under Settings, Brand appearance, confirming your organization holds the rights to display the mark. We show your logo and colors inside your workspace. Other brands appear as neutral monograms unless RateHighway holds an approved asset for them.

[STUB: important question, answer pending.] Owner: legal (OQ-36).

RateMonitor Elite is RateHighway’s pricing product. Agent Awareness is a standalone member of that family with optional integration.

Live price truth for the Price Freshness and Price Accuracy KPIs, and a read-only Agent Awareness dashboard tab inside RME. Account setup, prompts, evidence, team and billing stay in Agent Awareness.

No.

That is the plan. Your partner exchanges an organization-scoped embed key for a short-lived token and shows a read-only dashboard inside their screens. You create and revoke the key under Settings, Integrations.

[STUB: important question, answer pending.] Owner: partnerships (OQ-37).

[STUB: important question, answer pending.] Owner: legal.

Agent Awareness is a creation of project APOGEE. [STUB: important question, answer pending.] One public sentence is pending from its owner (OQ-38).

Use the suggestions address in the footer of every page. Suggestions are read by the product team.