SubIQ

Subscription intelligence for ServiceNow

The ServiceNow renewal number nobody can explain.

SubIQ is a free, read-only check that finds fulfiller licence drift — in custom tables, in inherited roles, and in seats nobody is using — before your renewal turns it into a negotiation.

Every year someone signs one. Fulfiller licences are consumed by activity, not by headcount, so your subscription count drifts — quietly, for years — while the org chart stays the same.

Then the renewal arrives, and the number is the number. You can query it, but you cannot correct the twelve months of drift behind it in the two weeks you have to respond.

Request the free check by email →

Who runs this?

If you have ever been asked "do we actually need all of these?" and had to answer from memory, this is for you.

Where does the drift come from?

Three places, and not one of them is anybody's mistake.

Custom table overlap. Extending core ITSM is normal, sensible platform work. What follows is not obvious: the extension can end up recorded as costing nothing while behaving exactly like the paid table it extends. Both things are true at once. Nothing in the platform surfaces the contradiction, so nobody has ever decided which reading is correct — because nobody has been shown that a decision is required.

Accidental role inheritance. People end up holding paid roles nobody chose to give them, because roles are inherited. Someone joins a group for an unrelated reason and arrives holding an expensive one. There was no decision, no approval and no record of intent — just a person who now counts against your subscription.

Activity is not access. Holding a fulfiller role is not the same as using it. Someone who signs in daily and never touches a ticket they did not raise looks identical to a working agent — if you go by sign-in dates. Sign-in dates are what most licence reviews go by, because they are what is easy to get.

Why has nobody caught it?

Not negligence. Three structural reasons.

The platform records what you bought in careful detail, and what you used barely at all. Custom tables are classified once, at creation, and never revisited. And the only people who could investigate properly are the same people keeping the service desk running.

So the drift is invisible until it is expensive. And the moment it becomes visible is the moment it becomes a negotiation.

Why does the month matter?

ServiceNow's financial year ends in December. Renewal and true-up conversations cluster in the months before it — which means the question "do you need all these seats?" tends to arrive in Q4.

Finding out is quick. Acting on it is not. Confirming that a specific person genuinely no longer needs a paid role means asking their manager, checking they are not mid-project, and agreeing a date. That is weeks of ordinary organisational work, per person.

Start in autumn and you have evidence and a cleaned-up count. Start when the renewal lands and you have an opinion.

What is included in the free check?

One read-only script. You paste it in, it prints a report, you read it.

It reports what it found and names what it could not see, so you can judge how far the answer reaches. No signup, no trial, no data collection, no obligation.

Request the free check

A person replies. There is no form and no mailing list.

What's delivered?

Three things, and you should know the shape of all of them before you email anybody. One is free and most people stop there. The other two are paid — a fixed number, quoted before any work starts. Nothing here is metered, nothing renews by itself, and there is no tier you get moved up to.

StageWhat you getWhat it costs
The free checkDiagnostic A read-only script you run yourself. It names the licensed tables, which of your own custom tables descend from them, and how many people hold a paid role with no evidence of doing the work it pays for. Free. No signup, no call, no obligation.
The full analysisResolution Named people rather than counts. For each one: whether there is evidence they have done the work the role exists for, where the role came from, and which of the two possible fixes actually holds. Delivered as a document with the evidence beside every claim. Fixed price, agreed before anything starts. One-off. It ends when it is delivered.
The standing watchPrevention Runs inside your instance. Compares what you are consuming against the number on your order form, and tells you as you approach it — and again if you pass it. Months before a Q4 conversation, not during one. Fixed price, quoted with the analysis. Nothing recurring for hosting: it runs on your instance, not on our servers.

Why the analysis is worth paying for, in one paragraph. Taking one person out of a group and removing the role where it originates look like the same fix on screen. They affect very different numbers of people, and only one of them holds — do the wrong one and the role quietly comes back. Working out which is which, person by person, is most of what the manual version of this costs, and it is why the going rate for doing it by hand runs into five figures.

Everything above stays inside your instance. The watch mails through your own system; we are not in the path and we never see your numbers.

You tell us what you bought. We do not read your contract, and we are not going to guess at it.

What will it not do?

It will not tell you that you are compliant, and it will not tell you that you are not. It reports exposure and shows its working. A human draws the conclusion.

It will not tell you to cancel anyone's licence. Reclaiming a seat that is genuinely in use is the worst outcome available here, so the check deliberately under-reports — it says worth reviewing where a louder tool would say safe to remove.

And it will not see everything. Where it cannot answer a question, it says so instead of guessing. That is the part most tools get wrong, and it is exactly the part that matters when you take the results to your account team.

It was built and verified against a current release. Releases differ, so if yours classifies things differently the check stops and tells you rather than reporting something misleading. Tell us which release you are on and we will confirm before you run anything.

How do you request it?

Tell us roughly how large your instance is and we will send the check, with notes on reading the output. If it misses something that matters to you, say so and we will extend it.

Start with the free check. Taking it no further is a perfectly reasonable outcome.