What your customers ask

best project management tool for remote teams?

Asked 12 Mar, 09:40 UTC · Example — not your data

Here’s what it read to answer that

asana.com
monday.com
clickup.com
Break found
veyra.com

Three sites were read. Veyra wasn’t one of them.

We find where your business is missing from AI answers.

What AI says about your business

Understand and improve how AI represents your business.

See what AI actually says about your business, get the pages built to address the gaps we find, then see what changed after you publish.

Free to run. If you want the pages and re-run, that’s $49 once.

  • Free
  • No signup
  • Usually ready in under a minute
Break found

What we found

We followed the break into the site itself.

Three findings stood out, each grounded in the site’s own content.

Finding 1

GPTBot is blocked from crawling the site

veyra.com/robots.txt · Read 12 Mar, 09:40 UTC · Example — not your data

1User-agent: GPTBot
2Disallow: /

These directives tell GPTBot not to crawl any page on the site. Pages it cannot crawl cannot become sources for answers it builds from the site.

Finding 2

The site does not clearly describe what Veyra offers

veyra.com · Read 12 Mar, 09:40 UTC · Example — not your data

“Veyra — Work better, together.”

The page gives the business a tagline, but does not clearly state what Veyra offers or which category it belongs to. That makes it harder for an answer engine to connect Veyra with questions about that category.

Finding 3

The FAQ does not explicitly connect each question to its answer

veyra.com/faq · Read 12 Mar, 09:40 UTC · Example — not your data

Question

Does Veyra work for remote teams?

Answer

Yes. Veyra is built for teams working across time zones.

The question and answer appear together on the page, but the page does not explicitly identify them as a question and answer pair. That connection can be made clearer for machines reading the page.

Three concrete gaps. We show what to change and help you publish it.

What AI reads for

Where is your company right now?

We look across your site for the information AI can read, understand, connect, and use when forming an answer.

Read

Can AI access and extract the important information?

Understand

Can it identify what the business is, what it offers, and what it is relevant to?

Connect

Can it connect questions, answers, products, pages and supporting information?

Cite

Can the resulting information become usable evidence in an AI answer?

Three states, one question

What separates these states is not how much a site publishes. It is how much of that information AI can read, understand, connect, and use when it forms an answer.

Hover a layer to see what that state means, and what opens it up.

From issues to exact changes

AI access

The site blocks GPTBot from reading its pages.

robots.txt

Access fix

Allow GPTBot to read the pages that matter.

robots.txt

Updated robots.txt

The corrected access rule, ready to publish.

robots.txt

Business meaning

The homepage does not clearly state what the business offers.

Homepage

Meaning fix

State clearly what the business offers and who it serves.

Homepage

Publish ready HTML

The corrected page content, ready to publish.

HTML

Disconnected answers

The FAQ does not clearly connect each question to its answer.

FAQ / Q&A

Connection fix

Connect each question to its answer, and link the pages that belong together.

FAQ / Q&A

Connected answers

Question-and-answer pairs and page links, ready to add.

FAQ / Q&A

Here’s what changes

When the gaps are fixed, your business can become easier for AI to cite.

CitePack turns the gaps it finds into specific changes you can publish.

Before

Veyra is missing from this answer.

What is the best A/B testing tool?

1.source-a.example
2.source-b.example
3.source-c.example

veyra.com

Not cited.

Changes

Access restoredrobots.txt
Meaning clarifiedHTML
Information connectedFAQ / Q&A

After — projected

Veyra appears as a source.

What is the best A/B testing tool?

1.source-a.example
2.source-b.example
3.veyra.comSource
4.source-c.example

Projected outcome. Not a guarantee.

The result depends on how an AI engine reads, interprets, and selects sources.

Inside the repair dashboard

See exactly what’s holding you back — and how to fix it.

CitePack turns the gaps in what AI can read into clear changes you can publish and check again.

Sample view

Cited in AI answers

Perplexity

Cited

ChatGPT

Cited

Claude

Cited

Google Gemini

Not cited

Four engines, asked separately. Each answer is recorded with the date it was given.

Changes published

Three of the four changes have been published. The last one is still open.

What changed over time

Each point is one recorded check, not a forecast.

What needs attention

  • AI readers blocked from the site

    veyra.com/robots.txt

  • Homepage does not say what the business offers

    veyra.com

  • Questions and answers not paired

    veyra.com/faq

Recently published

  • Access rule published12 Mar
  • Homepage wording published14 Mar
  • Question-and-answer pairs published18 Mar

Every check runs on all four

PerplexityChatGPTClaudeGoogle Gemini

No engine stands in for another.

One row per engine. What it said, and when.

Each engine is checked separately. We show what came back, which source was used, and what changed when the same question was checked again.

Same question, asked of each engine

Sample view
  • Perplexity

    Cited

    veyra.com/guide

    Now cites you

  • ChatGPT

    Cited

    veyra.com/compare

    Now cites you

  • Claude

    Cited

    veyra.com/faq

    Re-check scheduled

  • Google Gemini

    Not cited

    Another source

    Not cited in this check

What changes over time

Watch how your results change as you publish the fixes.

Every change gives you something new to check. See what changes across the AI engines over time.

Cited in answers, over time

Sample view

Each point is one recorded check on one date. An engine can change what it says without anything on your site changing, so the line moves both ways.

By engine

Perplexity

ChatGPT

Claude

Google Gemini

One row per engine, each asked separately. A filled mark is a check that came back cited.

What was published, and when

  • Access rule published

  • Question-and-answer pairs published

  • Homepage wording published

  • Product information connected

  • Links between pages improved

  • New page published

An example of the changes a report calls for. What yours asks for depends on what the diagnosis finds.

What we can promise

We can’t promise what an AI engine will say. We can show you what it actually said.

AI engines make their own decisions about what they read, interpret, and use. CitePack measures the result and shows you what changed.

We don't run the engines.

ChatGPT, Perplexity, Claude, and Google are independent systems. We ask them questions and report what comes back.

Sometimes there is nothing definite to report.

When an engine gives us no result we can verify, we show that clearly instead of filling the gap with an assumption.

You can check the same thing yourself.

The underlying pages are yours. CitePack gives you a structured way to inspect what the engines returned and what changed afterward.

What we measured ourselves

The answer doesn’t hold still.

up to 60%

of the sources an engine cites change between two identical questions asked the same day.

CitePack A/A ledger · 12 same-prompt pairs per engine · 2026-07-18

2 of 28

citations disappeared between two readings, two weeks apart, on seven real sites that changed nothing. The answer moved to someone else instead.

CitePack A/A ledger · 7 third-party sites, 28 engine-level pairs · 2026-07-17

Start fixing what’s holding you back

Your path from scan to a site AI can cite.

See what is missing. Prioritise what matters. Publish the changes. Check what changed.

  1. 01

    Scan

    CitePack checks your site and the four AI engines to find what is getting in the way.

  2. 02

    Prioritise

    See the issues that matter most and where to start.

  3. 03

    Fix

    Get specific changes and ready-to-publish outputs for the gaps we find.

  4. 04

    Check again

    Run the same checks again and see what changed.

The $49 report

The free scan shows you where to look. The $49 report gives you the work to do.

Get the full diagnosis, the specific fixes, and the pages ready to publish.

Free

  • Run the scan
  • See the initial findings

$49 one-time

  • Get the full report
  • The detailed diagnosis and publish-ready fixes
FAQ

Questions, answered.

CitePack asks real answer engines — ChatGPT, Perplexity, Claude, and Google (Gemini) — the question your customers ask, and shows you the answer they gave and which sites they named. If yours wasn't one of them, we read your own page and tell you what is missing from it and where it sits, down to the line. Then we write the pages your diagnosis calls for, and we ask the engines the same question again after you publish.

See what AI says about you. Then do something about it.