SkillsInventory.comSKILLS TRACKING PLATFORM
Running One Skills Inventory Across Multiple Sites or Plants
Industry & Role Playbooks

Running One Skills Inventory Across Multiple Sites or Plants

Rovaryn Digital· November 20, 2026· 9 min read

Why three plant spreadsheets add up to no inventory at all

A corporate operations director asks a simple question on a Monday morning: the Dayton plant has a customer ramp coming, and who across the other two sites is qualified on the new inspection step? The Dayton lead has a skills sheet. So do the other two plants. One uses a 1–5 scale, one uses "trained / not trained," and the third rates people as "A, B, C" with no written definition of what a B means. Two of the sheets have a column called "Forklift" and one has "PIT." Nobody can say whether a "3" in one building is a "Competent" in another, and the only way to answer the question is to phone three supervisors.

That is the real cost of site-by-site spreadsheets. Each file may be accurate for its own owner, yet together they can't be compared, rolled up, or used to borrow a qualified person from the next site.

You will walk away from this article able to run one skills inventory across multiple locations, with shared definitions, local owners, and coverage by site that you don't rebuild by hand each quarter.

Why the problem grows as sites are added

A single location can get by on one person's memory and a well-kept file, and a skills inventory across multiple locations is a different problem. Add a second plant and the file forks. Add a third and the forks drift: columns get renamed, a local manager adds a skill the others don't track, a sort breaks a formula.

The spreadsheet-risk research is blunt about this. Panko's work, summarized in Powell, Baker and Lawson's "Errors in Operational Spreadsheets" (2008), found that 94% of spreadsheets contain at least one error, with an average cell error rate of about 5.2%. A 2024 study reported by Phys.org (Poon et al., Frontiers of Computer Science) found the same 94% figure for spreadsheets used in business decision-making. Three independent copies of a matrix, each hand-edited by a different person, are three chances to be in that group, and the errors are invisible to anyone comparing across sites.

The labor market makes cross-site visibility more than a tidy-up exercise. Deloitte and The Manufacturing Institute (2024) found that 65% of manufacturers cite attracting and retaining talent as their primary business challenge, and that U.S. manufacturing could need as many as 3.8 million new workers between 2024 and 2033, with up to 1.9 million of those jobs potentially going unfilled. When qualified people are that hard to find, the person already on your payroll at the next site is the cheapest source of coverage you have, if you can find them.

Setting up a skills inventory across multiple locations

The structure that works is simple to state: one set of definitions, owned centrally, and one set of records, owned locally.

Central owns the vocabulary. Someone at the corporate or operations level maintains the skill list, the proficiency scale, and the certification categories. This is a small job once it is done properly, and it is the thing that makes every other comparison possible. If you are starting from nothing, begin with the skills that matter on the floor rather than a generic list. Our guide to a skills matrix for manufacturing walks through choosing the skills and rows worth tracking.

Local owns the ratings. The supervisor or area manager who sees people work is the right person to assess them. They enter and update ratings for their own site, and they are accountable for keeping them current. Central should not rate people it has never watched.

Everyone uses the same four definitions. Before any site enters data, agree on these in writing:

  1. Skill names. One name per skill. "Forklift," "PIT" and "Powered truck" become a single entry.
  2. Proficiency scale. A numbered scale with a one-line behavioral anchor at every level, so a 3 means the same thing in every building. A common choice is a 1–5 scale running from "awareness" to "can train others."
  3. Blank versus zero. A blank means "not assessed." A zero means "assessed and cannot do it." Mixing these silently distorts every coverage number downstream.
  4. Certification fields. For each credential, record the person, the date earned, the expiry date, and who recorded it.

A worked example of why the third rule matters: say a site has 20 operators and only 12 have been assessed on a skill, of whom 9 are competent. Treating the 8 unassessed people as zeros reports 45% coverage (9 of 20). Reading only the assessed people reports 75% (9 of 12). Both figures are real, and they mean very different things, so the report has to say how many people sit behind it.

Standardizing the scale before you compare anything

Standardizing skills across plants fails most often at the rating scale, not the skill list. Skill names are easy to reconcile in an afternoon. Ratings are not, because they encode each manager's private standard.

Run a short calibration before the first roll-up. Pick three or four skills, and have managers from different sites rate the same described scenarios. Where they disagree, rewrite the anchor until they would agree. Managers who have never compared notes tend to find they read a "3" differently.

Expect some friction. The strictest site will look weakest on paper, and local managers who rate hard may feel penalized next to colleagues who rate generously. Say up front that the first roll-up is a baseline, that the point is consistency, and that no one is being ranked on it. Then revisit the anchors after the first review cycle, once real disagreements have surfaced.

Reading skills coverage by location

Once definitions match, skills coverage by location becomes a question you can actually answer. For each critical skill, count how many people at each site are rated competent or better, then compare that count against the minimum you need per shift.

Three views pay for the setup:

  • Coverage by site. For each skill, how many qualified people does each location have? This is where a site carrying one person on a critical skill stands out immediately.
  • Single points of failure. Any skill held by exactly one person at a site is a risk. If another site has two or three, you have a candidate for loan, training, or a shared-coverage arrangement.
  • Cross-site borrowing. With a shared scale, "can anyone cover the second-shift inspection step at the Dayton plant?" becomes a filter, not a round of phone calls.

The gaps you find feed straight into planning. The cross-training matrix guide explains how to turn a thin-coverage skill into a who-trains-whom plan. If you want a workbook for that step, the Cross-Training Coverage Planner lets you set a redundancy target per critical skill and plan who-trains-whom, using your own matrix as the input.

Multi-location certification tracking without the audit scramble

Certifications are where inconsistency between sites turns into a compliance problem. Each plant may record forklift evaluations, for example, in a different place and with different fields. The record is not optional: under 29 CFR 1910.178(l), powered industrial truck operators must be re-evaluated at least once every three years, and the certification record must name the operator, the training date, the evaluation date, and the evaluator (EHS Careers, 29 CFR 1910.178). Refresher training can also be triggered earlier by accidents, near-misses, unsafe operation, or changes to equipment or the workplace.

Penalties for gaps are real and they adjust every year. OSHA's maximums, as adjusted effective January 15, 2025, are $16,550 per serious or other-than-serious violation and $165,514 per willful or repeat violation (U.S. DOL/OSHA). These are the 2025 amounts, not necessarily the latest. Because the figures change annually for inflation, confirm the current numbers with OSHA, and check the wording of 29 CFR 1910.178(l) against the regulation itself, before you rely on them. This is information, not legal advice. Verify your obligations with the relevant authority.

A single register with the same four certification fields at every site means one expiry report instead of three. Our piece on OSHA certification tracking covers building that register in more detail. One boundary to keep in mind with any tracker, spreadsheet or software: it records what you enter, including self-reported certification data and expiry dates. It does not verify a credential with the issuing body, so someone still has to check the evidence.

Keeping local ownership without losing control

Shared data raises a reasonable worry: if every site can edit, who stops a stray change from breaking the structure? And if only central can edit, you have rebuilt the bottleneck.

The answer is permissions and history. In a spreadsheet you have neither: no access control and no change history, which is why multi-site files tend to end up locked in one person's inbox. SkillsInventory is built around defined roles (Admin, Manager, Analyst, Viewer and Super Admin) and an append-only audit log, so site managers can maintain their own ratings while the shared definitions stay stable and every change is traceable. Our article on skills data access control and the audit trail covers how to decide who should edit what.

SkillsInventory ships with the O*NET taxonomy pre-loaded on first login, so every site starts from the same skill list rather than a blank page, and custom skills and bulk CSV import are available where your plants need their own entries. It is deliberately narrow: skills inventory, gap analysis and certification tracking. HRIS and LMS integrations are not in the current product, so rosters come in by CSV for now.

Where to start

If you are in one location or have fewer than about 50 employees, stay in Excel. A shared structure is overkill. If you run two or more sites and a shared skills inventory across multiple locations would help, and you can't currently answer a cross-site coverage question in under an hour, do this in order:

  1. Write the four shared definitions and get every site manager to sign off.
  2. Calibrate the scale on three or four skills across sites.
  3. Pull every site's data into one structure and mark unassessed as blank.
  4. Read coverage by site and certification expiry by site, then plan your first cross-training moves.

If you would rather start with a workbook, the Cross-Training Coverage Planner is a place to begin the coverage planning step. If you are ready to move off the spreadsheets, see how pricing works: it is flat-rate by tier, not per seat, so adding a site does not add a seat bill.

Ready to go beyond the guide?