A Metrc reconciliation routine that holds up.

Every inventory horror story I have heard starts the same way: the store found a big variance during a count, and by then nobody could say when it happened, who touched the product, or whether it was theft, a sync failure, or a typo at intake. The trail was cold because the store only looked when it had to.

Reconciliation is not a project. It is a rhythm. Here is the one I build with stores, broken into daily, weekly, and monthly layers, with one rule underneath all of it: every layer has a named owner.

Daily: close the loop on sales

Before the store closes out, one person confirms that what the POS sold matches what Metrc recorded. Most modern POS systems sync automatically, and most sync failures happen quietly. The daily check is short:

• Compare the POS sales count and units against Metrc's reported sales for the day.
• Clear any failed or queued transactions before close. Do not let them ride overnight.
• Log any manual adjustment with a reason. "Adjusted" with no note is a finding waiting to happen.

Same-day waste and damage logging belongs here too. A dropped jar gets documented the shift it breaks, with a photo, not remembered on Friday.

Weekly: cycle counts by zone

Full counts are too slow to do often, so most stores do them rarely, which means variances age for weeks. Cycle counting fixes the math: split the inventory into zones and count one zone per day.

• Monday flower, Tuesday pre-rolls, Wednesday concentrates, Thursday edibles, Friday everything else. Adjust to your menu.
• Count during slow hours. A rushed count creates the variances it is supposed to catch.
• Every zone gets touched weekly, so no discrepancy is ever more than seven days old. A seven-day-old variance is an investigation. A sixty-day-old variance is a write-off and a compliance risk.

Monthly: full count and a real audit

Once a month, count everything wall to wall and reconcile against Metrc. Then go one step further than most stores do: review the month's activity, not just the month-end number.

• Pull the month's adjustments and read them. Repeated manual adjustments from the same user or on the same product category is a pattern, and patterns have causes.
• Check for stuck packages: anything in transit too long, active packages showing zero quantity, or product approaching expiration.
• Verify every incoming transfer's manifest closed clean. Receiving errors are the most common source of phantom inventory.
• Write down what you found, even when the answer is "nothing." A documented history of clean reconciliations is exactly what you want to hand an auditor.

When you find a variance

Investigate while it is warm. Check the cameras for the affected zone, pull the package history, talk to the people who worked the shifts in question, and document the conclusion with the evidence attached. If the variance exceeds your state's reporting threshold, report it the way the rules require. A self-reported variance with a documented investigation reads completely differently to a regulator than one they find for you.

The routine sounds like a lot. In practice the daily check is ten minutes, a zone count is twenty, and the monthly audit is an afternoon. That is a small price for never having to explain a cold variance to the state.