Health digest
Sigil is invisible when it works, which is the point of the product and also a problem. An administrator who never opens the portal has no way of knowing whether 100% or 60% of their mailboxes are actually getting a signature.
The health digest is the answer. It takes what the portal already holds and mails it to your administrators, so the numbers arrive without anyone going to look for them.
What it contains
Section titled “What it contains”| Section | What it says |
|---|---|
| Coverage | What share of your mailboxes have ever applied a signature successfully, and up to five that never have |
| Apply failures | How many attempts Outlook did not accept, across the organisation |
| Directory gaps | Attributes your live templates reference that a meaningful share of mailboxes do not have |
| Banners closing | Any campaign banner whose window ends within a week |
| Waiting on approval | Drafts submitted for review and not yet decided |
| Booked publishes | Any scheduled publish not yet fired |
| Running rollouts | Any staged rollout in flight, and what percentage it is at |
Empty sections are left out rather than printed as zeroes.
While delivery is paused
Section titled “While delivery is paused”If delivery is paused, the digest says so before anything else, and the subject line leads with it rather than with a coverage figure. Every number below that line means something different when nobody is being served, and an administrator who reads the coverage first has already been misled by the time they reach the explanation.
The list of mailboxes that have never had a signature applied is left out for the same reason. Under a pause it is the whole organisation, and it describes a decision an administrator made rather than anything to fix.
Everything else still appears, and still counts as needing attention: a failing apply, a banner about to close and a draft waiting on an approver are all things worth fixing before delivery resumes, which is exactly when a pause is in force.
Mailboxes you have excluded from Sigil are left out of the coverage and directory gap figures entirely. They are deliberately not being sent a signature, and a recurring email that chased your administrators about them every week would be the fastest way to make the digest into something people filter away.
Directory gaps are scoped to what you actually use
Section titled “Directory gaps are scoped to what you actually use”The gaps section does not report on every attribute a template could reference. It reports on the ones your live templates actually depend on, including the fields that only appear as conditions on a section.
That is the difference between a useful line and noise. Nobody needs telling that
extensionAttribute12 is unpopulated if no template mentions it.
An attribute has to be missing on at least 10% of mailboxes before it is mentioned. Below that it is ordinary sparseness, and a handful of people without a mobile number is not something worth mailing anybody about.
A placeholder that is not a real field is reported too, since a mistyped placeholder resolves to nothing for everybody, which is exactly the kind of thing worth surfacing.
Attribute coverage in the portal is the fuller version of the same audit, and the place to go once the digest has told you there is something to look at.
A healthy week still gets a digest
Section titled “A healthy week still gets a digest”The digest reports rather than alerts. An organisation with nothing wrong gets a short one, with a subject line saying so.
This is deliberate. A message that only arrives when something is broken is an alert, and alerts do not tell you the product is working. “You are at 98%, nothing to do” is the message that makes the coverage number visible.
The subject line leads with the number that matters, so the digest is useful from the inbox list without being opened.
A running rollout and a booked publish appear in the digest but do not make it an attention-needed one. They are things you set up on purpose.
Seeing one before it arrives
Section titled “Seeing one before it arrives”You do not have to wait a week to find out what the digest says, or whether it sends at all.
| Button | What it does |
|---|---|
| Preview | Renders the digest your organisation would get right now, in the page. Nothing is sent |
| Send me one | Mails it to you alone, so you can see it land in a real inbox |
Both live in settings, next to the frequency.
Neither moves the schedule. Previewing or testing today does not delay the real digest, so there is no cost to pressing either.
Send me one goes to you and to nobody else. It is there to prove the delivery path works, and a test that mailed every administrator is a test nobody would run.
The preview is the same digest the scheduled send would produce, built by the same code rather than a separate rendering of it. A preview that showed something subtly different to the real thing would be worse than no preview, because it would be believed.
If your organisation is too new to be sent one, the preview still renders and tells you it would be held back. See below for when that applies.
How often it arrives
Section titled “How often it arrives”| Setting | Behaviour |
|---|---|
| Weekly | The default |
| Monthly | For organisations where weekly is more than they want |
| Off | No digest at all |
Change it in settings. Every digest carries a link to that page, so turning it down or off never means hunting for the switch.
It defaults to on because a digest nobody discovers is a digest nobody reads. It is a service email about your own organisation, with an off switch linked from every send.
Who receives it
Section titled “Who receives it”Every administrator of the organisation.
An organisation with no administrators yet, which happens partway through onboarding, is not sent one and is not marked as having received one. It gets its first digest once somebody is there to read it.
Organisations that are not being served are not sent digests either. That covers a suspension, a deletion grace window, and every billing reason signatures have stopped: a cancelled subscription, a trial that ended without a card, dunning that ran out, or a managed client whose provider’s own subscription has lapsed. None of them wants a weekly report about signatures nobody is receiving.
A cancellation that has only been scheduled is not one of those. Signatures are still going out until the day it takes effect, so the digest carries on until then and stops with them.
When it does not arrive
Section titled “When it does not arrive”A brand new organisation with no directory read and no activity at all is skipped until there is something to say. A mail full of zeroes reads like a fault.
If the directory could not be read on the day a digest was built, the digest is still sent but omits the coverage section rather than reporting 0%, which would be alarming and wrong.
Every organisation due a digest is sent one on the same run. Sigil used to work through the list in batches, so a busy day could push a digest to the following morning; it no longer does, and no organisation waits its turn behind another.
A digest is not sent to an organisation whose signatures have stopped, whether that is a suspension, a pending deletion or a billing reason. The test is the same one the signature path applies, so the digest stops when the signatures do rather than on a rule of its own.
Where the numbers come from
Section titled “Where the numbers come from”Coverage counts a mailbox as covered once the add-in has reported applying a signature successfully at least once. That is evidence Outlook actually took the signature, rather than evidence Sigil served one, and the two are not the same thing.
The same telemetry drives the activity view, which is where to go for per-mailbox detail. The digest is a summary of it, not a separate measurement.
What it is not
Section titled “What it is not”It is not an alerting system. There is no threshold you can set, nothing pages anybody, and a digest is not sent early because something broke.
If a signature stopped working an hour ago, the digest is the wrong tool. Open activity.