Free tier — 1 competitor, tracked weekly

Every claim your competitors make. Extracted, scored and watched.

Paste a URL. Get every number, positioning line and footnote as structured rows — and an alert the day one of them changes. No rollout, no sales call, no noise about a moved footer.

Start free See what it pulls out

No credit card First claims back in minutes No annual contract

app.claimspy.io/c/crayon.co

crayon.co 94 claims · 3 changed in 7d Sorted by relevance All types

88

“Our win rate increased from 16% to 45%.”

A ZoomInfo product marketing lead is quoted saying their win rate rose from 16% to 45% after adopting Crayon.

Testimonial Case study 16% → 45% win rate crayon.co/
74

“Spring 2026 Grid Leader — Recommended · High Flyer · Challenger · Leader”

Claims G2 named them a Leader in its Spring 2026 Grid for competitive intelligence.

Position read from a picture crayon.co/
74

“Customers see 40% higher win rates on average.”

Metric Across customers 40% win rate crayon.co/
41

“Monitors your competitors and alerts you to relevant intel.”

Capability crayon.co/product

90 more claims Last crawled 4h ago Next run in 3d Every capture time-stamped

01How it works

Three steps, and none of them is a call.

Up and running before a legacy vendor would have finished qualifying you.

Paste a competitor's domain

The entire setup. No tracker script, no CRM connection, no onboarding call.

We open the whole site and read it

We open every page the way a visitor does, click open whatever is hidden, and turn what we find into sortable rows — each one scored and linked back to the spot it came from.

Get told when something moves

A metric going 40% to 51% reaches you. A restyled button does not.

02Why it sees more

Most of a competitor's site never reaches an AI agent.

Three reasons: how many pages it opens, whether they finished loading, and whether anything got clicked. Same competitor, same morning — several times the material.

It reads a few pages. We read the domain.

Point one at a competitor and it follows a handful of links, then stops. We follow every link off the home page, then every link off those, until the domain runs out.

Where a chat agent gets to — 3 pages Where ClaimSpy gets to — all 150

Depth 0 The page you handed it 1
Depth 1 Everything linked from the home page 8
Depth 2 Product pages, plans, comparisons 41
Depth 3 Customer stories, changelog, the fine print 100
Followed to the end, every run 150

It gets a blank page. We get the real one.

Most sites arrive empty and finish building themselves in the browser. Grab one the quick way and you get the empty version, or a bot check.

What an agent gets

crayon.co

Verify you are human

crayon.co needs to review the security of your connection before you continue.

crayon.co

You need to enable JavaScript to run this app.

Home · Privacy · Terms

0 claims

What ClaimSpy gets

crayon.co

Win more deals with competitive intel

G2 Leader — Spring 2026 Grid Rated 4.6 / 5 across 412 reviews

“Customers see 40% higher win rates on average.”

“$6 Million in pipeline influenced.”

12 claims

Same address, a second apart. One has words on it you can hold a competitor to. The other is a bot check and an apology.

It never touches the page. We click through it.

Half a competitor's boldest copy is folded behind a “read more”, a tab, or a slide nobody presses — which is usually the half worth having.

crayon.co/product

“Leader — Spring 2026 Grid for Competitive Intelligence”

4.6

“Rated 4.6 out of 5 across 412 verified reviews”

3yr

“Named a Challenger three years running”

How it works

“Cut research time by 80%^† and win more deals.”

† Internal survey of 42 customers, 2023. Results vary.

What's included

“Unlimited competitor tracking on every plan.^*”

* Fair use applies. See terms for limits.

Security

“SOC 2 Type II certified since 2024.”

Report available under NDA on request.

Integrations

Data & privacy

Service levels

Support

Changelog

Every slide, then every drawer. On their site that is 5 sliders and 68 drop-downs — and what is folded inside them is usually what they assume nobody will find.

03What it catches

Then it takes each claim apart.

Finding the copy is half of it. The other half is keeping the pieces that let you argue back — the number on its own, the small print underneath, and the date it last changed.

One real claim, and everything we keep about it.

The headline is the flex. The small print is the truth. Your counter-copy needs both, with a date on them — so we store them as separate, searchable pieces rather than as a paragraph you have to re-read every month.

What is on their page

Cut research time by 80%^† and win more deals.

^† Based on an internal survey of 42 customers conducted in 2023. Individual results vary. Not a guarantee of performance.

The big line is the promise. The small grey line is what actually backs it up — and the one a summary drops.

What ClaimSpy keeps

Claim Cut research time by 80%

Type Metric Across customers

The number 80 % reduction research time

Small print n = 42 · self-reported · 2023 · “results vary”

Found in crayon.co/product — behind a closed “How it works” drop-down

History Changed 3 times since Mar 2026 — used to say “by 60%”

Proof Time-stamped copy of the page, 30 Aug 2026

7 pieces you can sort, compare and quote

04Signal, not noise

One week of a competitor's site. Two very different inboxes.

Most tools alert on change. ClaimSpy alerts on meaning — so it is quiet most weeks, and worth reading when it isn't.

What ClaimSpy sent 3 events

crayon.co Value changed 2d ago

40% 51%

“Customers see 51% higher win rates on average.”

Metric Across customers crayon.co/

klue.com New claim 4d ago

“The only competitive enablement platform built for the enterprise.”

Position new · homepage hero klue.com/

crayon.co Dropped 6d ago

“$6 Million in pipeline influenced.”

Metric gone from 4 pages crayon.co/customers

Three emails in seven days, each one a claim that moved. You read all three.

What a legacy tool sent 47 events

Hero background image swapped2d

Nav links reordered2d

14 CSS class names changed3d

Cookie banner wording updated4d

Blog post published: “5 CI trends for 2026”5d

Testimonial carousel now autoplays6d

Footer copyright year bumped6d

Pricing page meta description edited7d

Not one of these changed what the competitor claims. ClaimSpy saw all 47 and sent nothing — that is the feature you are paying for.

05For legal & brand protection

A screenshot is not evidence. This is.

The week you challenge a claim is the week it comes off their site. We already sealed a copy, stamped by an authority nobody in the argument controls.

Independently time-stamped

An RFC 3161 authority attests this record existed at Sun, 30 Aug 2026 08:50:08 GMT. Alter any part of it and the stamp stops verifying.

Retrieved: 30 Aug 2026
Method: Automated browser, whole page loaded
Page state: Collapsed sections expanded before reading
Viewport: 1280×900px
Authority: timestamp.digicert.com
Position: Append-only chain, record 1,284
Digest: 2402a7c6ffd9b9c2d1cc3a50afeb88b8b2de9e787fb3bd28651514fb29894698

An independent authority, not us

The stamp is an RFC 3161 authority's, not our clock. We cannot backdate a record. Neither can you — that is the point.
One seal over three things

Page image, extracted claims, collection record — one digest. Change one, it breaks.
Method stated up front

How it was fetched, at what viewport, what was clicked open. Volunteered, that is methodology. Discovered later, it is the end of your exhibit.
Honest about its own gaps

An unstamped capture says so. So does a highlight we cannot place with confidence.

For false-advertising complaints, comparative-advertising disputes, NAD and ASA challenges, demand letters. Admissibility is your counsel's call — be wary of any vendor who says otherwise. We control the record, not the ruling.

06vs. the alternatives

An enterprise suite is too much. A chat prompt is too little.

ClaimSpy does one job properly and charges accordingly.

ClaimSpy compared with enterprise competitive intelligence suites and general AI assistants
	ClaimSpy	Klue, Crayon & co.	ChatGPT + a URL
Time to first result	Minutes, self-serve	Weeks — demo, pilot, onboarding	About a minute
Price	Free, then $39–$299/mo	$15k–$60k+/yr, annual contract	~$20/mo
Reads sites that build themselves in the browser	Opens the page in full	Yes	Often returns an empty shell
Depth of collection	Whole site · clicks drop-downs, tabs and sliders · reads words in pictures	Whole site, plus news and social	One URL, at rest, never clicked
Remembers last month's wording	Full claim history	Yes	No persistence between chats
Tamper-evident record of the page	RFC 3161 stamp + sealed digest	Screenshots inside the platform	None
Maps fine print to the headline	Core of the extraction	Not a first-class object	Usually summarised away
Output	Sortable, scored rows → CSV, Notion, Slack	Cards inside their platform	Prose you re-format by hand
Alerts on restyles and layout tweaks	Never — claims only	Constantly	n/a

Competitor pricing reflects publicly reported ranges; neither publishes list pricing. The rest compares product shape, not quality — an enterprise CI programme is a real thing to run, just not what ClaimSpy is for.

07Pricing

Priced like a utility, because that is what it is.

Start free on one competitor. Move up when weekly stops being fast enough.

Free

One competitor, watched properly. Not a teaser.

$0forever

Start free

Includes

1 competitor domain
Weekly crawl
Up to 50 pages per run
Full extraction, typing & relevance scoring
Change alerts by email

Starter

The shortlist you actually lose deals to.

$39/month

Start free

Everything in Free, plus

3 competitor domains
Up to 150 pages per run
2 seats
CSV & XLSX export
Weekly email digest

The ones worth answering.

Why not just ask ChatGPT to read the page?

For a one-off look, do that. Make a habit of it and three things break: it reads the handful of links it happens to follow, it grabs the page before it has finished loading, and it never clicks anything — so every claim behind an accordion, a tab, a carousel or an image stays invisible to it. Section 02 puts all three side by side against a real run.

And it forgets. A chat cannot tell you what the page said in March; here every claim keeps its history, so next month's wording arrives as a diff rather than a re-read.

How fast do I see something after signing up?

The first crawl starts immediately and the claim list fills in over the next few minutes. The change feed needs a second run — a diff needs two sides — so on the free tier that is a week later. That first alert is usually the moment it clicks.

What counts as a “page”?

One page opened in one run, counted per domain. We prioritise pages where claims live — home, product, pricing, case studies, comparisons — so 50 pages goes further than it sounds on a site with a 400-post blog.

Can it see pages behind a login or a paywall?

No, and it will not try. ClaimSpy reads the public marketing pages any visitor sees. No credentials, no gated content.

Is crawling a competitor's site above board?

We read public marketing pages at a polite rate, with a crawler that identifies itself and honours robots.txt. It is what your sales team already reads by hand before every competitive call — read consistently, and remembered.

Is a ClaimSpy record admissible in court?

Not ours to promise, and be sceptical of any vendor who says it is — admissibility turns on your jurisdiction and your counsel, not a feature list. What we control is the record: an RFC 3161 stamp from an independent authority, one digest over the image, the claims and the collection record, a position in an append-only chain, and a written statement of method. Unstamped captures say so.

Why that beats a screenshot you took yourself: yours is an assertion — your clock, no proof it has not been edited. A record collected on a schedule before the dispute existed is the part you cannot arrange afterwards.

What if nothing changed this week?

You hear nothing. A tool that manufactures an update to look busy has trained you to ignore it. Your claim matrix is still there when you want it.

Do I have to talk to anyone to buy this?

No — not on Free, and not at $299. Card checkout is still being wired up, so a paid upgrade is one email we action the same day. If you want a human, hello@claimspy.io reaches one, and it will not be a rep.

Pick one competitor. See what they are actually claiming.

Free, takes a URL, and you will know in minutes whether it is worth a line item.

Launch ClaimSpy See pricing

No credit card · No sales call · No annual contract

Every claim your competitors make. Extracted, scored and watched.

Three steps, and none of them is a call.

Paste a competitor's domain

We open the whole site and read it

Get told when something moves

Most of a competitor's site never reaches an AI agent.

It reads a few pages. We read the domain.

It gets a blank page. We get the real one.

It never touches the page. We click through it.

Then it takes each claim apart.

One real claim, and everything we keep about it.

One week of a competitor's site. Two very different inboxes.

A screenshot is not evidence. This is.

An independent authority, not us

One seal over three things

Method stated up front

Honest about its own gaps

An enterprise suite is too much. A chat prompt is too little.

Priced like a utility, because that is what it is.

The ones worth answering.

Pick one competitor. See what they are actually claiming.