← Selected work

Consumer trust · scam detection

CheckIfScam

One box. Any scam. An instant, plain-English verdict.

A trust-first scam detection platform that turns uncertainty into a clear answer in seconds.

Client
CheckIfScam.com
Industry
Consumer trust · scam detection
Year
2025
Role
Founder · Product, UX, design system, content & SEO architecture
Visit live site (checkifscam.com)
checkifscam.com
CheckIfScam — flagship screen

The core idea

One box. Any scam. An instant, plain-English verdict.

The product is shaped around a single moment: the second a user pastes something in. Everything else exists to give that verdict somewhere to lead.

CheckIfScam verdict result for a Royal Mail delivery scam

Verdict UX

The answer first. Then the why. Then the one thing to do next.

Platform scale

12,000+
Threats tracked
5,415
Scam messages indexed
3,755
Flagged domains
2,386
SEO landing pages
378
Recovery guides
106
UK brands verified

Overview

A consumer trust platform shaped around the second a suspicious message arrives. One input, a calm verdict, and — if it's already too late — somewhere to go next. Free, anonymous, UK-focused.

The problem

01

Scam tools were fragmented.

Existing tools each solved a slice of the problem — domain reputation here, screenshot analysis there, reporting after the fact. Nowhere could an anxious person paste almost anything suspicious and be told, clearly, what it was and what to do next.

The insight

02

Users don't want a tool. They want an answer.

In the seconds after a suspicious text arrives, no one needs more interface. They need somewhere measured and credible that hands back a verdict in plain English. The whole product is shaped around that emotional truth.

Panic Mode — guided five-step recovery flow

Panic Mode

A guided recovery journey for users who have already clicked or paid. One decision at a time. No blame.

The strategy

03

One input. A wider system behind it.

A universal input sits at the centre. Behind it, recovery guidance, brand verification, live intelligence and a structured scam library give the verdict somewhere to lead — and let the platform scale through depth, not surface pages.

The design

04

Calm, plain, quietly confident.

Sage green over alarm red. Editorial spacing over dashboard density. Confidence shown without theatre. Nothing in the interface raises the temperature; the whole point is to lower it.

Brand verification page for HSBC

Brand register

Verified UK brand pages double as a public trust register — what's real, what scammers impersonate, how to tell.

The system

05

Built on a structured taxonomy.

A scam model — brand, channel, tactic, asset, stage, recovery — powers every surface, held to quality gates that reject thin or low-confidence pages by design.

The system

Six surfaces. One platform.

Each pillar is a self-contained product. Each one feeds the others.

01

Universal input

One field for text, links, numbers, charges, emails and screenshots — routed through a hybrid AI verdict pipeline.

02

Verdict UX

Plain English over threat-intel jargon. A clear answer, not a risk score.

03

Panic Mode

A five-step recovery journey for users who have already clicked or paid.

04

Brand register

A public register of verified UK brands — what they look like, what scammers impersonate, how to tell.

05

Live scam radar

Trending scams, flagged domains and emerging tactics, read from the platform's own signal in real time.

06

Structured library

A taxonomy of every scam pattern — the spine of the platform's discoverability.

Positioning

Held together where the category ships in pieces.

Most scam tools take on a single slice — a domain check, a screenshot scan, a report form. CheckIfScam carries the user from suspicion to verdict to recovery without ever handing them off.

Universal input across formatsHeldA slice
Plain-English verdictHeldRisk score
Guided recoveryHeldAbsent
Verified brand registerHeldAbsent
Live scam intelligenceHeldA slice
Free, anonymous core journeyHeldMixed
CheckIfScamElsewhere

On the phone

Built for the moment something arrives.

Most scams reach people on a phone, mid-task. The mobile experience is the product — sticky actions, recovery-first flows, and a verdict that fits a single thumb-scroll.

checkifscam.com

CheckIfScam

Is this a scam?

Paste anything suspicious — message, link, number, charge — get a clear answer.

MessageLinkPhoneCharge

Trusted by 12,000+ checks

The universal input — one field, every accepted format.

Result

!

Likely scam

92% confidence

This message impersonates Royal Mail. The link does not match the official domain.

What to do next

  1. 01Don't tap the link.
  2. 02Forward the message to 7726.
  3. 03Delete it from your inbox.
View recovery guide →

The verdict — reassurance, evidence, a single next step.

In use

Across the surfaces.

Live Scam Radar showing trending scams and flagged domains

Live intelligence

Trending scams, flagged domains and emerging tactics — read from the platform's own signal in real time.

Scam Library search interface

Scam library

A structured library indexed by category, brand and tactic — the spine of the platform's organic discoverability.

Global search across brands, scam alerts and recovery guides

Global search

One search spans every surface. The user never has to know which part of the platform holds the answer.

"
Not another cyber tool. The calm friend you'd ring when something didn't feel right.
Joseph Hilson — Founder, CheckIfScam

My role

Founder · Product, UX, design system, content & SEO architecture.

  • 01Named, positioned and shaped CheckIfScam as a calm, UK-focused alternative to alarmist scam tools.
  • 02Authored the universal-input thesis and the three-mode intent split: received, already clicked, checking for someone else.
  • 03Designed the verdict-led UX — reassurance, then evidence, then a single next step.
  • 04Built Panic Mode as a guided five-step recovery journey — mobile-first, no blame.
  • 05Created the full design system: tokens, components, and bespoke verdict, risk and recovery patterns.
  • 06Modelled the scam taxonomy spanning brand, channel, tactic, asset, stage and recovery.
  • 07Architected a quality-gated programmatic SEO system that rejects thin pages by default.
  • 08Set up the hybrid AI verdict pipeline with human-readable explanations and calibrated confidence.

In summary

Trust isn't signalled. It's built — into the pace of the interface, and the confidence of the answer that comes back.