The Morning Exception Report: a Tekmetric daily report with four lines and a rule for each

This is the guide behind the video: the four lines, the Tekmetric report each one comes from, the exact rule to write for each, and how to get the numbers out of Tekmetric without anyone exporting a file by hand.

The four lines

1. Yesterday against normal. Sales, car count and average repair order next to the same weekday in recent weeks.
2. The quiet estimates. Every estimate still unapproved after three days, oldest first, with the advisor's name.
3. Declined work. Work from the past week that a customer declined and nobody followed up, like a set of brakes a customer said no to on Tuesday.
4. Tech efficiency. Sold hours over clocked hours, flagged only when a tech falls well below their own average.

The agent only reads. It writes nothing back to Tekmetric, so a wrong number costs you a minute, not a repair order. Think of it like a pre-trip inspection: you write down only what is wrong, not every tire that is fine. On a normal day the report is short.

Who this is for

Owners and managers of Tekmetric shops who know the numbers are in there and stop opening the reports after the first busy week. If you run several locations, the same four lines work per shop, side by side. For the bigger picture of how reporting automation works with Tekmetric, see our Tekmetric reporting automation guide.

What Tekmetric already gives you

You do not need a new report. Tekmetric's own write-up lists an End of Day Report with shop-wide sales and profit for a date range, Service Writer and Technician reports, a Technician Hours Report, a Declined Jobs Report and more, and it defines technician efficiency as sold hours divided by clocked hours. The real-time reporting page adds car count, ARO (average repair order), close ratio and labor hours. What neither page describes is a scheduled report that lands in your inbox, so the habit of opening them is on you. That habit is the part to automate.

Line by line: the source and the rule

Copy the rule column as written and change the numbers to fit your shop. The thresholds below are starting points, not benchmarks.

Line	Tekmetric source	Rule to write	Shows when
1. Yesterday against normal	End of Day Report, real-time reporting (sales, car count, ARO)	Compare yesterday with the average of the same weekday over the last four weeks	Any of the three is more than 20% off that average
2. Quiet estimates	Repair orders in estimate status	Estimate sent, not approved, older than three days	Always, oldest first, with advisor name and amount
3. Declined work	Declined Jobs Report	Declined in the last seven days and no follow-up noted	Always, largest first
4. Tech efficiency	Technician Hours Report (sold hours, clocked hours)	Yesterday's sold hours over clocked hours against that tech's own 20-day average	More than 15 points below their own average

Why "their own average" on line 4: comparing one tech with another punishes the person who gets the hard diagnostic jobs. Comparing a tech with their own normal shows a bad day, which is what you can act on.

What the report looks like

This is an illustration of the layout, not a result from a shop:

Morning Exception Report, Tue 7:00 AM	Example line
1. Yesterday against normal	Car count 14 against a usual 18 for Mondays; sales and ARO normal
2. Quiet estimates	RO 8838, $1,240, 4 days, Priya; RO 8829, $610, 6 days, Sam
3. Declined work	RO 8841, front brakes, declined Tuesday, no follow-up
4. Tech efficiency	Nothing flagged

Four lines, two of them action items. A normal day might have one.

Getting the numbers out of Tekmetric

There are two routes, and you can start with either.

Webhooks. A shop admin can add a webhook under Settings, then Integrations, which sends real-time appointment and repair order updates as they happen. That covers line 2 and line 3 well, because estimates and declines show up as repair order changes. The agent keeps its own small history so it can compare yesterday with recent weeks.

The API. A third-party review of the Tekmetric API says access is not self-serve: it needs an active paid subscription and manual approval that typically takes two to three weeks, with a cap of 25 records per query. That reviewer sells managed API layers, so treat it as one view and confirm the details with Tekmetric before you plan around it.

Ready-made tools exist too. AutoRx, for example, sends a daily summary by email, text or app with 15+ KPIs pulled from Tekmetric. If a fixed KPI list fits you, that may be enough. The Morning Exception Report differs in one way: it shows only what broke a rule you wrote, so most mornings it is short.

Setting it up in two weeks

1. Write your four rules in the table above, in your own numbers.
2. Pick the route: webhooks if you want to start this week, the API if you already have approval.
3. Choose where it lands: one text or email to the owner and the service manager, before the first customer arrives.
4. Run it next to your normal routine for two weeks. Open the Tekmetric reports as usual and check whether the report caught what you found.
5. Tune the thresholds. If a line flags every day, the rule is too tight. If it missed something you found, add a rule.
6. Decide who acts on each line. A quiet estimate goes to the advisor on it; a declined job goes to whoever makes follow-up calls.

What to watch out for

• Write access. The agent should only read. Sending a follow-up to a customer is a person's decision, made after reading the line. Cortex's defaults are on our trust and security page.
• Too many lines. Six lines that never flag train you to ignore the report. Keep four until one earns its place.
• No history. "Below normal" needs a few weeks of stored daily numbers. Start storing on day one.
• Comparing techs with each other instead of with their own average.

How to tell it worked

After two weeks, the report has caught every quiet estimate and declined job you would have found by hand, most mornings it has two lines or fewer, and each flagged line has a named person who acted on it. Our daily briefing page shows the same idea across several systems, and building it for your shop is part of the free AI Operations Audit.