How to cut estimated meter reads: a playbook for copier dealers

October 19, 2026 · 5 min read

On this page
e-automate ships with a meter source called Auto-estimated, a built-in code used when the system has to batch-estimate a meter during contract billing. It sits in the same list as Contract, Fax, Phone and Web Suspect: ECI built a specific code for the moment an actual read doesn’t show up in time to bill on.
Here’s roughly how it works, and what you can do about it before it happens.
Why e-automate reaches for an estimate in the first place
When you enter a meter reading, e-automate checks it against what it expected and warns you if the reading fails that check. To build the expected number, it works through three methods in order. First, if the equipment has three months of usage history, it uses the average monthly usage. Second, if that history isn’t there yet, it falls back to the manufacturer’s stated volume for the meter, if someone entered it. Third, if neither is available, it can attempt an estimate from whatever usage data does exist, using this formula: Estimated Meter = Last Meter + ((Monthly Average / 30.25) x Number of Days since Last Meter).
If there’s no usable history and no manufacturer volume on file either, none of the three methods produces a number, and e-automate has nothing to check a reading against, so it can’t even warn you one looks wrong. In practice, that’s more likely on newly added equipment or a contract that was just onboarded, before a manufacturer volume or enough usage history is on file.
The playbook
- Configure the validation buffer. In equipment options, the meter validation buffer sets how far above the known average a reading can run before e-automate issues a warning instead of just accepting it. e-automate’s own example: a meter averaging 1,500 copies a month with a 50% buffer gets accepted up to 2,250 before it warns. Worth knowing too: for less than 31 days since the last known reading, e-automate uses a different, more forgiving version of that math. The documentation notes separately that usage isn’t always constant (a company might do the bulk of a month’s printing on a single day).
- Request readings before the overage cycle ends, not after. The Meter Reading Request task in e-agent can email, fax, or build a phone list to chase a reading ahead of the deadline. Set it 5 days out on a September 15 to October 14 billing period, for instance, and it starts sending on October 9, then, depending on how the task is configured, can keep sending until a usable read comes back.
- Feed device data in directly where you can. FMAudit integration syncs meters collected from the devices themselves into e-automate, helping eliminate the redundant manual entry that meter billing otherwise involves. When properly configured, FMAudit also populates a Last FMAudit custom property with the Last Meters Audit date each time it updates a meter, a date your team can check when investigating a device that seems to have gone quiet.
- When someone has to estimate a meter by hand, use the console’s estimate function instead of a guess. In the Meter Reading Console, selecting Estimated as the source enables an Estimate button that populates an estimated reading when there’s enough history and your settings allow it. If the system can’t estimate the reading, it explains why, and a person can still enter one manually and check the box identifying it as an estimate.
- Report meter sources separately. Auto-estimated, Estimated and Web Suspect (an e-info reading that came in outside meter expectations) are three different things. Keeping them apart in your own reporting makes it easier to see how many of the meters you billed last quarter were actual reads versus estimates.
It’s worth reviewing these settings when you onboard new equipment or a new contract, when the history these methods depend on may not yet be on file.
What we hear from dealers on this
In how dealers we talk to rank the jobs they want off the service desk, meter reads come first, ahead of dispatch and call intake, ahead of toner. Dealers also raise a related concern: whether an AI agent will post a bad meter read. We built Auralis so nothing posts outside the threshold a dealer sets. A read that falls outside it is routed to a person instead of posting automatically. Your team stops keying reads in but still decides what gets billed. Auralis ingests readings from shared inboxes and writes them into e-automate through our ECI partner integration, via the ESN API.
If you want the mechanics, How do we connect to your E-Automate? lists the six e-automate objects Auralis reads and writes, the path it uses, and the five steps from first call to live.
Sources
- ECI, e-automate online help (v250), Creating Meter Sources.
- ECI, e-automate online help (v250), Additional Info - Meter Volume Checks.
- ECI, e-automate online help (v250), Setting Equipment Options.
- ECI, e-automate online help (v250), Using the Meter Reading Console.
- ECI, e-automate online help (v250), Configuring Meter Reading Requests.
- ECI, e-automate online help (v250), Integrating FMAudit with E-automate.
- Auralis product information and conversations with dealers (internal).
For e-automate dealers
See Auralis on your own service desk
A 30-minute walkthrough on your workflow — meter reads, dispatch, supplies and escalation.
Book a demo
Data Scientist
Related posts

Month-end meter-read close: a checklist for office technology dealers
Contract billing in e-automate runs through four tabs, and its validation warnings show up in the billing queue. A six-step checklist for closing meter reads at month end.

The dealer tech stack in 2026: e-automate, DCAs, CEO Juice, portals
FMAudit, ESN, PIP, e-info, MobileTech and CEO Juice, described the way ECI's help pages and CEO Juice's own site describe them: what each piece of an e-automate dealer's stack does.

Office technology dealer M&A due diligence: a service operations checklist
A dealer's pre-close review can stay financial and legal. Here is a six-step checklist for the service side: devices outside your radius, meter collection, call-type data, volume and what to automate.