15 questions people ask AI before hiring an immigration consultant, and what answers them
I took 15 real questions Indian visa applicants ask an AI assistant before hiring a consultant, and checked whether 18 real immigration firm homepages actually answer them.
Before someone calls a visa consultant, more and more of them ask an AI assistant first: am I even eligible, what will this cost, how long will it take. That conversation happens whether or not a consultant's website is built to be part of it, and increasingly the AI's answer, not the consultant's homepage, is the first impression a prospective client gets.
I took the same 18 Indian immigration and visa consultant homepages I scanned on 4 August 2026 with my public AI-readability engine and mapped them against 15 real questions applicants ask before choosing a consultant. Here is what an AI can and can't currently lift off these pages, and what would fix it.
The 15 questions, and the fix for each
State eligibility criteria in plain, structured text on the homepage (age, education, work experience, language score), not buried in a downloadable brochure.
If the criteria aren't in text on the page, an AI can't check them for you.
Publish the current points threshold or a simple points breakdown directly on the page, and update it when it changes.
A number stated once beats a number the AI has to guess at.
List a realistic processing timeline by visa type, not a vague 'varies' or 'contact us for details.'
Vague timelines get skipped; specific ones get quoted.
Show a clear fee structure or fee range and what it covers (application, documentation, appeals) instead of 'contact for a quote.'
No visible price means no visible answer.
Lay out the pathways you handle (skilled worker, family, student, investor) with a short who it's for line for each.
A menu of pathways is easier for an AI to match than a paragraph of prose.
If you track and can stand behind a success rate, state it plainly along with the basis for the number.
An unstated success rate is invisible to an AI, even if it's excellent.
State your registration number and registering body clearly in the page text, not just as a small logo image.
A logo in an image isn't text an AI can read; the same claim in words is.
Publish your refusal or refund policy in plain text, even if the honest answer is 'no refund' or 'conditional refund.'
Applicants ask this before they ask about price. If it isn't stated, they assume the worst.
Publish or link to a current occupation or NOC list relevant to your main destination countries, ideally as text, not a PDF.
A PDF is a dead end for most AI engines; a text list isn't.
State clearly whether post-landing services (job search, settlement support) are included or offered separately.
If a service isn't described on the page, it doesn't exist to an AI answering on your behalf.
Publish a short, honest comparison of the destinations you handle (Canada, Australia, and others) and the kind of profile each suits.
A page that only pitches one country reads as biased, not helpful.
State the minimum language band or score required for your main visa pathways directly in text.
This is one of the most searched questions in the category; it should be one of the easiest to answer.
Answer dependent inclusion directly for each visa pathway you handle, rather than leaving it to a consultation call.
Family questions are decision-stage questions; answer them before the call, not during it.
List the documentation and proof of funds support you provide as a concrete, named service, not a general 'we help with everything.'
Specific services get matched to specific questions; vague ones don't.
State your process for handling rejections or delays (appeal support, reapplication, communication updates) in plain text.
This is a risk question. An AI that can't find an answer will tell the applicant that risk is unaddressed.
What I found scanning 18 immigration firm sites
Scanning the same 18 sites gave a median score of 91.5, the strongest vertical I have measured to date, with 14 of the 18 scoring an A.
15 of the 18 sites publish structured data, with a median score of 92; the 3 that don't collapse to a median of 53. FAQ schema is on 7 of the 18, the highest adoption I've seen in any vertical, but still under half.
The pattern under all 15
The firms that answer these 15 questions well tend to state the answers in plain, structured text: eligibility criteria as a list, points thresholds as a number, fees as a range, a registration number spelled out.
The firms that miss bury the same answers in blog posts, downloadable brochures or vague marketing prose, none of which an AI engine can reliably lift into a direct answer.
Frequently asked questions
Can an AI assistant actually tell me if I'm eligible for a visa?
Only as well as the consultant's website lets it. If a site states eligibility criteria in plain text, an AI can check you against them. If the criteria are buried in a brochure or a call, the AI has nothing to work with.
How do I know if a visa consultant is actually registered (ICCRC, MARA)?
Look for the registration number stated in text on their site, not just a badge image. In this scan, that kind of detail was one of the clearest gaps: information that exists but isn't written where a person or an AI can read it.
Why do some visa consultant sites answer these questions and others don't?
It usually comes down to structure. Sites with JSON-LD structured data and FAQ schema scored far higher in this scan; sites without them, even with the same information, scored much lower.
Should I trust an AI's answer about a visa consultant over the consultant's own website?
Treat it as a starting point, not a final answer. An AI is only repeating what it could find and read on the page, so a well-documented site gives you a more reliable AI answer, and a thin site should make you ask more questions directly.