ITISYOUVerify

How it works

Three steps, and the setup work each one really needs

This page is the long version. Nothing here is a summary of a feature we have not built. If a step sounds like work, it is work.

How is this different from my automation's own error alerts?

Your automation platform's error alerts tell you when the automation itself threw an error. They say nothing about whether the CRM record or email that came out the other end is actually correct: a workflow that runs top to bottom without an error can still write to the wrong record, skip a step, or send to the wrong address, and none of that raises an alert. We don't watch your automation at all. We read HubSpot and Resend back ourselves, independently, and check what's actually there against the rules you set.

Do you ever modify my CRM or resend my emails?

No. We only ever read. We do not create or edit CRM records, and we do not send a replacement email on your behalf. If a run fails, fixing it is still on you.

Three steps

What setting this up actually involves

  1. Connect HubSpot and Resend

    Give us read access to your HubSpot contacts and to the Resend account your acknowledgement emails go through. We never ask for write access, and we never fetch a URL you supply: only the two connected providers.

  2. Define the expected result

    Tell us which HubSpot property carries your correlation value, and set the checks that must be true: the record exists with the right value, the email reached the right recipient, the status reached the point you need. Your automation sends us one signed message per enquiry naming what it expects: that message is a trigger, never proof on its own.

  3. Receive the evidence

    We read the record and the email status back ourselves and tell you verified, failed, unverified or pending, with the reason for each check and the evidence attached. No fifth status, no silent guessing.

What you need before day one

A HubSpot account you can grant us read access to, a property on your HubSpot contact records that carries a correlation value for each enquiry, a Resend account whose message events we can read, and a small change to your existing automation so it sends us one signed event per enquiry. This is real setup work, not a one-click connection. There is no separate onboarding guide yet: the how-it-works page is the full instructions until one is written.

What we report

Status: Verified

Every check you set as mandatory has enough supporting evidence from HubSpot and Resend.

Status: Failed

The evidence contradicts one of your mandatory checks, or the completion deadline passed while we could still reach both providers.

Status: Unverified

We could not get a clear answer: access, correlation or evidence was missing or ambiguous. This is not a pass and not a failure.

Status: Pending

Still inside the agreed completion window. We are still checking.

What this does not do

It does not watch your automation platform

We never see inside n8n, Make, Zapier or whatever runs your workflow. We only see what exists in HubSpot and Resend afterwards, so we cannot tell you which step of your automation went wrong.

It does not fix anything

We do not create or edit CRM records, and we do not send a replacement email. We report what we find; your team still does the fixing.

It does not cover every workflow

Version one checks exactly one shape: an enquiry that should create the correct CRM record and trigger an acknowledgement email, using HubSpot and Resend only.

It does not detect a run that never started

We only learn about an enquiry when your automation sends us an event. We cannot tell you that a run never started, because we have built no way to find enquiries your automation never reported. If it stops calling us we show nothing, not a pass, not a failure, and every result names the coverage we actually have.

It is not instant

Results depend on a scheduled check, not a live push. A result can take up to an hour to settle.

Questions people ask first

Can you tell me when a run never started?

No. We only find out about an enquiry when your automation sends us a signed event; if it never sends one, we have nothing to check, and we show nothing rather than treating silence as a pass. We have built no way to find enquiries your automation never reported, and we would rather name that blind spot than sell you a setting that does not close it. Every result shows the coverage we actually have next to it.

What does "coverage mode" mean?

It tells you how a workflow finds out about a run. There is one coverage mode today, "customer triggered", and it means we only see an enquiry when your automation tells us about it, so we cannot see a run that never started at all. It is shown next to your results so the blind spot is never out of sight. A second mode, where we would find the enquiries ourselves, is on the roadmap and is not built; it is not offered, because offering it would mean promising to spot enquiries we have no way to see.

What if the evidence source is down?

If HubSpot or Resend is unreachable when we try to check, we don't guess and we don't report a pass. The run shows as unverified with the reason recorded, and we retry automatically within a bounded number of attempts. A missed deadline only counts as failed if the evidence sources were actually reachable at the time; a provider outage never turns into a false failure.

Is this real-time?

No. We check evidence on a schedule, not instantly. A result can take up to an hour to settle, depending on the completion window your workflow is set to.

What's the difference between an email being "accepted" and being "delivered"?

"Accepted" means Resend, the sending service, took the message. "Delivered" means the receiving mail server actually took it in. Those are different claims and we never merge them: a rule that requires delivery is not satisfied by acceptance alone. We also never treat an email being opened as proof anyone read it.