TRACKITWEEKLY FIELD GUIDE

Why Are My Locations Ordering So Differently When They Sell Similar Amounts?

Short answer: when locations with similar sales order very differently, the gap is usually structural, not sloppy. The locations likely have different delivery days, slightly different menus or portioning, different par levels written by different managers, counts taken at different times, or losses happening at one site that the others do not have. Any one of those differences will produce different orders from identical sales. The job is to separate the legitimate differences from the accidental ones, standardize the accidental, and leave the legitimate alone on purpose.

What this problem looks like in real life

  • Two similar locations differ by 20 percent or more on weekly order size for the same items.
  • One location constantly stocks out while another quietly overstockes the same products.
  • Each location's manager defends their numbers as correct, and each is partly right.
  • Comparing locations feels impossible because everyone counts and orders differently.
  • The differences persist across manager changes, which means they are built into the system, not the people.

The likely root causes, and how to tell them apart

1. Delivery schedules differ, so safety stock differs

A location that gets three deliveries a week can run lean. A location on a weekly drop needs a deeper buffer to survive the same demand. If orders differ but days-of-supply on hand are similar, this is your cause, and the difference is legitimate.

2. Menus and portioning are not actually identical

"Same menu" drifts. One location's kitchen uses a little more of an item per plate, or sells more of a particular dish because of the neighborhood. Small menu differences compound into real order differences. Compare usage per item, not just order totals.

3. Par levels were written locally, with local judgment

If each location built its own order-up-to levels over the years, you are comparing a set of personal systems. Different risk tolerances, different supplier relationships, and different past stockouts all got baked into the numbers.

4. Count timing and method differ

One location counts Sunday night after a busy weekend, another counts Monday morning after receiving. One counts in cases, another in units. The orders differ because the inputs differ, even when the shelves are identical.

5. One location is losing product the others are not

Spoilage, over-portioning, or theft at a single site inflates its usage and therefore its orders. The tell: that location's usage is higher than sales justify, consistently, in a way menu differences cannot explain.

6. Local supplier terms push different behavior

Different case minimums, different reps, different substitute habits from regional distributors. Location B orders more of an item because its supplier only sells it in bigger packs, not because it uses more.

How to figure out which cause applies

  1. Compare usage per item, normalized by sales, not order totals. Divide each location's weekly usage of an item by its weekly sales. Legitimate causes one and two show up as explainable ratios. Cause five shows up as a ratio that has no operational explanation.
  2. Map delivery calendars side by side. Different frequencies explain different order sizes instantly, and legitimately.
  3. Compare the two locations' par sheets line by line. Wide unexplained gaps with similar sales confirm cause three.
  4. Check count day and time at each site. A one-day difference between a post-weekend count and a post-receiving count can shift numbers by double digits on fast movers.
  5. Walk both locations' back rooms in the same week. Overstock at one and bare shelves at the other, on the same items, is cause three or five in physical form.

Practical corrective actions

  • Build one par framework, with local multipliers. Set company-wide baseline levels from combined usage, then adjust per location with documented multipliers for delivery frequency and known local demand. The framework is shared; the multipliers are deliberate.
  • Standardize the count across locations. Same day of week relative to delivery, same units, same method. Consistent inputs make location comparison meaningful for the first time.
  • Review locations side by side weekly. Ten minutes looking at two columns of the same items changes behavior faster than any policy, because nobody wants to be the unexplained outlier.
  • Investigate the location whose usage outruns its sales. Do it supportively: watch portioning, check waste habits, audit counts. Most single-site losses are process issues, and they are fixable once they are visible.
  • Normalize supplier terms where you can. Consolidate slow movers to one distributor with uniform case sizes, or negotiate matching terms across locations for your top items.

Where TrackItWeekly fits: multi-location visibility is its home ground. Every site runs the same weekly count workflow, rolls the same three-week usage average, and works from PAR levels and multipliers you can set centrally and tune locally. Color guidance and velocity visibility let a district manager see at a glance which location is drifting, and count and order history keeps every site's decisions on the record. Where it does not fit: it will not tell you why one kitchen portions differently. That still takes a visit and a conversation, with the numbers in hand as the starting point.

What to track going forward

  • Usage per item per location, normalized by sales.
  • Order size per location, and the documented multipliers that explain differences.
  • Days of supply on hand at each site, so delivery frequency differences stay visible and deliberate.
  • Variance between locations on shared items, trended monthly.

What not to do

  • Do not force identical orders. Identical orders on non-identical delivery schedules produce stockouts at one site and overstock at the other. Standardize the framework, not the totals.
  • Do not assume the outlier location is badly run. The outlier is often the one honest counter, or the one with a genuinely different supplier situation. Investigate before you standardize.
  • Do not compare locations on order dollars alone. Dollars hide everything: pack sizes, prices, and mix. Compare units and usage ratios.
  • Do not let each location keep its private spreadsheets. Private tools make private systems. Shared visibility is what turns five locations into one operation with five doors.

FAQ

How much order difference between locations is normal?

Enough to cover delivery frequency and real local demand differences, which is often 10 to 15 percent on order size. Beyond that, you should be able to point at a documented reason for every point of the gap.

Should district managers set all par levels themselves?

Set the baseline centrally, then tune with the local managers. Central-only levels miss local reality; local-only levels recreate the inconsistency you are trying to fix. The multiplier approach splits the difference on purpose.

What if one location's higher usage is just theft?

It happens, but it is the last explanation to reach for, not the first. Rule out count errors, delivery timing, portioning, and waste first. If the gap survives all of those, then investigate loss with proper care.

Do seasonal changes apply per location or company-wide?

Both. A shared seasonal multiplier keeps the company aligned, and a local adjustment layer handles a location near a seasonal event the others do not have. Track both separately so each is visible.

How often should we re-baseline the par framework?

Twice a year for most businesses, and always after a major menu or supplier change. Between re-baselines, only the multipliers should move, and each move should have a note.

Conclusion

Different orders from similar locations are not automatically a discipline problem. Delivery schedules, menu drift, local pars, count timing, single-site losses, and supplier terms each produce legitimate-looking differences, and your job is to sort the legitimate from the accidental. Build one par framework with documented local multipliers, standardize the counts, and review the locations side by side every week. The gap will shrink to what it should be: the real, explainable difference between your locations, and nothing more.