We show our working.

What the score is: and is not

The number is an educational estimate. It describes how a profile reads against criteria that the authorities publish themselves. It is not a prediction of what will happen, and it is not advice about your case. Real decisions are made by trained officers weighing evidence and impressions that no questionnaire can capture, including an interview, the quality of your documents, and factors specific to your history.

A high score is not a promise of approval. A low score is not a bar on applying. What the score does reliably is show you which parts of your profile carry the most weight, and which of your own answers sit at the weak end of that scale.

The factors and their weights

These percentages are ours, not the Home Office’s. The factors come from the published rules. The numbers beside them do not. UK Visas and Immigration does not publish a weighting for the genuine visitor assessment, and caseworkers do not score applications out of 100. These weights are our own reading of which factors the published criteria emphasise, and a different reader could reasonably choose different ones.

We publish them anyway, because a number you can argue with is worth more than a number you cannot see. They sum to 100 and that is enforced in code: a rule pack whose weights do not total 100 fails at build time rather than shipping a quietly wrong score.

Band thresholds

The score sorts into three bands. The thresholds below are our starting point and will be tuned as real outcome data arrives, if you come back and tell us what actually happened, that is what improves them.

• Green · Well positioned, 80 to 100. Strong on paper. What remains is the risk of an avoidable slip, a missing document or a careless inconsistency, rather than anything fundamental.
• Amber · Present with care, 45 to 79. A genuine trip carrying real presentation risk. This is the band where preparing the evidence properly actually changes the picture.
• Red · Not yet, 0 to 44. On these answers the gaps are structural, they sit in your circumstances rather than your paperwork. A report or a cover letter would not close them, so we will not sell you one.

Cross-checks: where your own answers are compared

One factor is not scored from any question. Information consistency comes from 9 deterministic comparisons run between the answers you gave. Your monthly income against your annual income, the cost of the trip against your savings, your travel dates against the nights of accommodation you have arranged, the date you are due back at work against the date you leave the UK.

This is the reason the check asks for approximate figures rather than bands. Two bands cannot contradict each other; two numbers can. Where they do, we show you both sides and say which fields they came from. A cross-check finding is never framed as a risk of refusal. It is a factual or documentation point to settle before you submit, and it says nothing about which version is the correct one.

A second factor, evidence readiness, asks a different question from the rest: not whether something is true, but whether you can produce a document for it. An application is assessed on what was submitted, so the gap between a sound account and an evidenced one is worth measuring on its own.

Ceilings: when the score is not the whole story

A weighted average has a flaw: a serious problem can be diluted by strong answers elsewhere. Without a correction, someone with a previous overstay and a refusal could be told they are well positioned, which would be both wrong and harmful.

So certain answers cap the band outright, whatever the points say. A removal or refusal of entry, an earlier misrepresentation on an immigration application, or staying with someone who has no lawful status each cap the reading at red. Each is a point at which we say plainly that a self-assessment is the wrong tool and a regulated adviser is the right one. A previous overstay, a criminal conviction, more than one refusal, applying while out of work, savings that do not cover the trip, a residence permit expiring within six months, or a stay longer than the route permits each cap it at amber. When a ceiling applies we say so on the result, explain why, and tell you what the points alone would have read. We would rather show you the lower number.

Why red profiles are not charged

If the model puts a profile in the red band, the weaknesses are structural. They sit in circumstances, not in presentation. No cover letter closes a gap in ties or funds. Selling a report in that situation would be taking money for something that cannot help, so we do not. The honest read is free.

Where the criteria come from

Every factor traces back to a published source:

• Immigration Rules, Appendix V: Visitor The genuine-visitor requirement and the rules governing this route.
• GOV.UK, Standard Visitor visa Eligibility, permitted activities and what evidence is expected.
• Immigration Rules, Part Suitability: grounds for refusal The general grounds, previous breaches, deception and criminality.
• GOV.UK, check if you need a UK visa The official check for whether your nationality requires a visa or an ETA.
• Home Office, immigration system statistics Quarterly entry clearance data, including visitor outcomes by nationality.

What we deliberately do not collect

The questionnaire asks about your circumstances, never for identifiers. We never ask for a passport number, a bank account number, a home address, a date of birth, your name, or your employer’s name, and there is nothing to upload. A tool that never holds those cannot leak them.

We do ask for approximate amounts and dates, because the cross-checks above are impossible without them. An amount is not an identifier: “about $2,000 a month” describes a very large number of people and singles out none of them. Every answer, figures included, stays in your browser. The assessment runs on your own device, and your answers never leave your browser unless you buy a report. We do count that a check was started and finished, with the band and a score range rather than the exact figure, and nothing else. See privacy for exactly what is counted and what happens when you buy.

Limits you should know about

• A questionnaire cannot capture evidence quality, interview performance, or officer discretion.
• A cross-check tells you two of your answers disagree. It cannot tell you which one is right, and it is not a judgement about your circumstances.
• The model does not know your nationality-specific context, and refusal rates vary widely between countries.
• Thresholds are our current best judgement, not a validated statistical model. We will say so plainly until real outcome data says otherwise.

Try the free check