For the complete documentation index, see llms.txt. Every page on this site is also served as Markdown: append `.md` to any URL, or send `Accept: text/markdown`.
Skip to content

Measure · Attribution

Every change to your pages, matched and measured

Attensira reads your pages on a schedule and records every change it sees: one it shipped, one you made from its advice, or one nobody suggested. It matches the change to the suggestion behind it, then re-reads the same prompts at +7, +14 and +30 days and reports moved, too early, or nothing it can prove yet.

7 days free · card requiredHow the trial works

Pages · attensira.com/alternatives/profoundattensira.com
Versions seen
30 Sep 2026 · first capturedHTTP 200 · JSON-LD present
1,657 words
2 Oct · changedJSON-LD absent: the server returned a near-empty page
36 words
4 Oct · changedBody and JSON-LD back
1,693 words
Copywriter agentNew wording · waiting
Say self-serve versus demo-first in the opening lines

Filed 30 Sep for “Attensira vs Profound”, where this page is already the top source ChatGPT cites. Once someone pastes it and marks it shipped, that prompt is re-read at +7, +14 and +30 days.

ChatGPT · 52% of citations · n = 18 answers

Re-reads booked: none. Two changes were seen, but no suggestion for this page has shipped, so nothing is measured and nothing is claimed.

Attensira.com's own Profound alternatives page on 5 Oct 2026. Attribution records every version it sees, including a broken deploy, and books no verdict until a suggestion ships.

Analyst agentAnalystattensira.com workspace, read over the Attensira MCP ·
  • Analyst agentAnalyst · Matches each change to the work behind it and reads the verdict, with its n.

What does the crew do with this reading?

The documented flow, step by step. It is an example: no brand, no IDs and no numbers.

See how the crew fixes this: Content optimization
“Did the change to this page move anything?”Example
  1. Read the question on every engine your plan tracks, once a day per country.
  2. Saw a new version of the page and matched it to the suggestion or delivery behind it.
  3. Kept every answer word for word, with its sources, as the evidence.
  4. Checked the result against the last window before calling anything a change.
  5. Filed what it found on the prompt's row, and in your Inbox when it needs a decision.
  6. Handed to Content optimization: the page that should have been quoted.
Last stateHanded to Content optimization
Example business · not a customer

How does attribution work?

Seeing a change, matching it and measuring it are three separate steps, and none of them claims the change caused anything.

  1. See the change

    Your pages are read weekly, and daily while a suggestion for them is open. A new version is recorded when the title, description, headings, structured data, status or body text changes past the noise.

  2. Match it to the work

    The change is compared with every suggestion and delivery on file for that page: shipped by us, applied suggestion, close to suggestion, field flagged, draft published, or unattributed.

  3. Re-read the prompts

    The prompts it targets, or else the prompts whose answers cited the page, are re-read at +7, +14 and +30 days. Each compares a week before the change with a week around the date.

  4. Decide once

    At +30 days the answers are pooled into one before and one after, the significance test runs once, and the label is stored: moved, too early, or nothing we can prove yet.

How is a change matched and measured?

  • See every change on your pages

    Pages are read weekly, and daily while a suggestion for them is open. A new version is recorded when the title, headings, structured data, status or body text changes past the noise.

    Page changes in the docs
    Page changes · one page
    1. Title changed
    2. Structured data added
    3. Whitespace only: below the noise
    Example business · not a customer
  • Match it to the work behind it

    The change is compared with every suggestion and delivery on file for that page, and labelled by how it got there.

    Page changes in the docs
    Matched as
    • Shipped by us
    • Applied suggestion
    • Close to suggestion
    • Unattributed
    Example business · not a customer
  • Re-read the same prompts

    The prompts it targets, or else the prompts whose answers cited the page, are re-read at +7, +14 and +30 days.

    Before and after in the docs
    Re-reads
    +7 days
    Booked
    +14 days
    Booked
    +30 days
    Verdict
    Example business · not a customer
  • Decide once, never say because

    At +30 days the answers are pooled into one before and one after, the test runs once, and the label is stored: moved, too early, or nothing we can prove yet.

    Before and after in the docs
    Verdict
    • Moved
    • Too early
    • Nothing we can prove yet
    Example business · not a customer

What does a page's baseline look like before a change ships?

Attribution compares each page against its own before. These are the answers that linked four of our own pages in the last 30 days; a change to any of them is judged against this count at +7, +14 and +30 days.

answers that linked / (www)
72 of 4,518
n=4,518 answers
answers that linked /compare
61 of 4,518
n=4,518 answers
answers that linked /alternatives/profound
58 of 4,518
n=4,518 answers
answers that linked /compare/attensira-vs-profound
47 of 4,518
n=4,518 answers
Analyst agentAnalyst · Holds the baseline per pageattensira.com workspace, read over the Attensira MCP on 4 Oct 2026 · ChatGPT and Perplexity · 24 of 31 days sampled · to

Nothing ships without your yes, unless you let a lane ship on its own.

One workspace setting decides how far the crew goes on its own. Draft is where every workspace starts.

  1. Ask

    Stops and asks before anything leaves the workspace. A card shows exactly what it wants to do.

  2. DraftDefault

    Does the work inside the workspace on its own and files everything else for your yes. Nothing is published or sent until you approve it.

  3. Act

    Carries safe changes through on its own: structured data, meta tags, alt text, internal links and heading levels. This is how you let work ship on its own.

  • Claims always wait

    A number, a price, a product or competitor name or a commitment waits for your yes in every mode, Act included. A change you edited before approving always counts as one. In the docs

  • Unanswered approvals expire after 7 days

    If nothing is decided in that window, the run closes without doing the thing and files a row in your Inbox saying so. In the docs

  • Every act has a receipt

    One dated row per act, with its status. Failed attempts are listed, not hidden, and a withdrawn one is marked cancelled, never removed. In the docs

  • You can take it back

    A pull request merges only when you press Merge, or under auto-merge for safe changes if you switched it on. A pull request is a proposal: closing it undoes it. A task that stays inside the workspace is undone by cancelling its session. In the docs

How the Inbox worksWhat agents can and never do

How is this measured?

What the numbers on this page stand on, from the docs.

Cadence
Each prompt, at most once per engine per country per day. The verbatim answer is kept. Docs
Samples
1x on Starter, 3x on Growth and 3x on Business per reading. Docs
Engines
ChatGPT, Perplexity, Claude and Google AI on every plan; Gemini from Growth. Docs
Markets
Up to 5 countries per workspace, topic or prompt. A country an engine cannot be read from is shown as not measured, never as zero. Docs
What a mention is
An answer that names your brand. A link to your own domain is a citation, counted on the same answers. Docs
Every number
A count with its n beside it. No blended visibility score. Docs
After work ships
Re-read at +7, +14 and +30 days. “No change” is a verdict.
What we do, and what we don't claim
What we doWhat we don't claim
Report the change and its measurement as two separate facts.That a change caused a move. It never says because.
Show counts without a percentage below thirty answers.A rate from a sample too small to hold one.

What does it cost?

On every plan · trial on Starter

Reading attribution spends no credits: opening a dashboard, filtering a chart or exporting numbers you already have is free on every plan. Credits pay only for agent work: chats, automation runs and ask calls.

What a piece of work spends scales with how much work it actually does, and Costs in the app shows what the month has cost.

Starter

$129/mo
Prompts
50
Credits a month
3,000
Engines
4
Sampling
1x per reading
Start free

7 days free · card required · $30 of credits (3,000 credits)

See plans and credits

Proof on our own sites

Sites we own · not customers

attensira.comChatGPT + Perplexity

Named in 2 of 1,481answers to the 17 questions that don’t name attensira.com, 2 Oct to 1 Nov 2026 (30 of 31 days sampled). With our own name in the question too: 310 of 2,274.

Firsts
15 firsts. The latest on a question that names no other product:“best AI search visibility tools for agencies”ChatGPT · linked to us ·
The tool’s own wordsChatGPT linked to attensira.com for the first time since we started reading this prompt on 15 September 2026, in an answer to 'best AI search visibility tools for agencies' on 25 October 2026.
Each is a first since we started reading, not proof our work caused it.
Work
8 drafted. Where they stand:
  • 3put live on the site
  • 1live in the workspace
  • 4approved, waiting on us to put live
Re-read, 3 of 3 put live: Nothing we can prove yet
attensira.com's own workspace · read 1 Nov 2026 · n=1,481 answers
modeldiplomat.comChatGPT + Perplexity

Named in 195 of 602answers to the 7 questions that don’t name modeldiplomat.com, 2 Oct to 1 Nov 2026 (30 of 31 days sampled). With our own name in the question too: 595 of 1,276.

Firsts
7 firsts. The latest:“best AI research assistant for debating”Perplexity · linked to us ·
The tool’s own wordsPerplexity linked to modeldiplomat.com for the first time since we started reading this prompt on 1 October 2026, in an answer to 'best AI research assistant for debating' on 16 October 2026.
Each is a first since we started reading, not proof our work caused it.
Work
No receipt yet
modeldiplomat.com's own workspace · read 1 Nov 2026 · n=602 answers

Shown as a baseline: nothing here is offered as a result of our own work.

Where else does the before-and-after show up?

The same verdicts feed every place that reports on work, so no page can claim more than the re-reads proved.

Where else does the before-and-after show up?
GoalsOne number from a fixed catalog, a target and a date. Its start is measured, never typed, and it reads ahead, on pace, behind or unmeasured every morning. Shipped changes sit beside it as markers, never as its cause.
ReceiptEvery act with its status: shipped, waiting, failed or cancelled. It says no change is proven until a re-read comes back moved, and shares on a private link that lasts 30 days.
Morning briefUp to three lines under “Changes we saw on your pages”, moved first.
Client reportsUp to ten changes matched to our work in the period, worded as their evidence card words them.

What does attribution not claim?

A re-read shows that two windows differ. It cannot rule out a model update or a competitor's new page in the same weeks.

  • It never says “because”, “thanks to” or “drove”. The change and its measurement are always two sentences.
  • Seeing a change never marks a work item shipped or approves anything. A person confirms that.
  • Pages are read as the HTML your server returns. Text that appears only after JavaScript runs is not seen.
  • Below 30 answers in a window, the card shows the counts and no percentage.
  • Page changes are in the dashboard only, not yet on the REST API or the MCP server.

What is the change being measured?

The one that shipped: here, an answer moved to the first sentence of a page. From the day it merged, the same prompts are re-read at +7, +14 and +30 days, and the verdict says moved, too early, or nothing proven yet.

Pull requestledgerly/website · #512

Answer “Does Ledgerly send VAT-compliant invoices?” first

attensira2 checks passed

content/features/vat-invoices.mdx

  1. removed: Invoicing that keeps up with you.
  2. added: Yes. Ledgerly adds the VAT number, rate and reverse-charge note that EU invoices need.
Web engineer agentWeb engineer · Opened the changeExample · Ledgerly, a fictional brand

What else do people ask about attribution?

Is this in the trial?
Yes. Pages are read and every change is recorded, including changes you make yourself. The trial includes $30 of credits (3,000 credits), so the crew drafts real fixes for you to approve; nothing ships without your yes. Re-reads run on whatever changed.
What does it cost in credits?
Reading attribution spends no credits: opening a dashboard, filtering a chart or exporting numbers you already have is free on every plan. Credits pay only for agent work: chats, automation runs and ask calls. What a piece of work spends scales with how much work it actually does, and Costs in the app shows what the month has cost.
Does anything change on my site without me?
Nothing ships without your yes, unless you let a lane ship on its own. Draft is the default: nothing leaves the workspace until you approve it. In Act, safe changes such as structured data, meta tags, alt text and internal links go ahead on their own, and anything that says something about your business waits for your yes in every mode. A pull request merges only when you press Merge, or under auto-merge for safe changes if you switched it on.
How is this measured?
Each prompt is read at most once per engine per country per day, and every rate carries the number of answers it was counted over. Sampling is 1x on Starter, 3x on Growth and 3x on Business. A change is reported only when it clears a significance test, and work that shipped is re-read at +7, +14 and +30 days.
Does Attensira take credit for changes it did not make?
No. Every change carries how it was matched, from shipped by us to unattributed, and a change nobody suggested says so. No sentence about a page change uses “because”, “thanks to”, “led to” or “drove”.
What does “moved” mean on a page change?
The pooled before and after for the change's prompts cleared a two-proportion significance test at 95%, with at least 30 answers in each window. It is observed movement, not proof that the change caused it.
How do goals fit in?
A goal is one number from a fixed catalog, a target and a date, such as named in 30 of 60 prompts on ChatGPT by 31 December. Attensira measures where it starts, reads it every morning and says whether it is ahead, on pace, behind or unmeasured. A number it could not read is never a 0.
What if I make the change myself?
It is still seen and measured. A change you made from our suggestion is matched as an applied suggestion, one close to it as close to suggestion, and a change nobody suggested as unattributed. Each gets the same re-reads.

See what changed, and what the answers did after.

The moment the trial starts, every prompt you track is read on every engine your plan includes, and you watch the rows fill in. The first report lands once half the reads are in, or after twenty minutes. After that, every prompt is read once a day.

7 days free · card requiredHow the trial works

Start free

7 days free · card requiredHow the trial works