Facilities Management Spreadsheet: Setup and When to Move On
A facilities management spreadsheet works for a small team when it is built like a database: one tab per record type (assets, work orders, preventive maintenance, vendors, leases), one row per thing, a unique ID in the first column, drop-down lists instead of free text, and one owner who controls the single master copy. It stops working when several people edit it at once, when you need to prove who changed a row and when, or when the same building lives in two lists that disagree.
Nothing about that is a knock on spreadsheets. They are free, everyone knows them, and plenty of facilities teams run on one for years. The problem is that a spreadsheet keeps no history and checks nothing on its own, and research on real spreadsheets in real companies shows how quietly errors build up when nobody is checking.
What a facilities management spreadsheet should track
Keep each record type on its own tab, and link the tabs with IDs rather than by retyping names. This is the layout we would start from:
| Tab | One row per | Columns that matter |
|---|---|---|
| Locations | Building, floor or room | Location ID, building, floor, room, use |
| Assets | Piece of equipment | Asset ID, location ID, make, model, serial, install date, cost, warranty end |
| Work orders | Request or job | WO ID, date opened, location ID, asset ID, priority, description, assigned to, date closed, labor hours, parts |
| PM schedule | Recurring task on one asset | Asset ID, task, frequency, last done, next due |
| Vendors | Contractor | Vendor ID, trade, contact, insurance expiry, contract end |
| Critical dates | Lease or contract deadline | Document, clause, date, notice window, owner |
A few rules make the difference between a list and a record:
- IDs, not names. "AHU-3" in a work order row should match exactly one row on the Assets tab. Names drift ("RTU 2", "Rooftop unit #2", "roof AC"), and then no filter can find every job on that unit.
- Drop-downs for anything you will filter on. Priority, status, trade and building should come from a fixed list using data validation, so "Urgent", "urgent" and "URG" never become three categories.
- Real dates in date columns. A date typed as text cannot be sorted or used to calculate the next due date.
- Never overwrite history. A work order gets a closed date, not a deleted row. A PM gets a new "last done" only after the old one is logged somewhere.
- One master file, one owner. Copies emailed around become versions, and versions disagree.
Our guides on physical asset inventory and preventive maintenance schedules cover what belongs in the asset and PM tabs in more depth.
How often spreadsheets are wrong
More often than the people who built them believe. Researchers at Dartmouth's Tuck School of Business applied a formal auditing protocol to 50 operational spreadsheets and, in their 2009 paper, "found errors in 0.9% to 1.8% of all formula cells, depending on how errors are defined." That sounds small until you remember that one wrong cell can feed every total below it.
The same team then measured what the errors actually did. In a second study of 25 spreadsheets from five organizations, they found 381 potential issues, and the developers confirmed 117 of them as errors. Nine of the 25 spreadsheets had no errors at all, 47 errors had no effect on the results, and among the rest, the published version reports that seven "exceeded $10 million." The largest single error was over $100 million.
The causes they heard from the people who built the spreadsheets will sound familiar to anyone who has inherited a facilities workbook:
- Time pressure, the reason cited most often
- Organic design, a file "inherited from a predecessor spreadsheet" or one that grew "without ever consciously being designed"
- Changing specs, and a lack of testing: "most of them did no testing as such"
Ray Panko's 2015 review pulled together 85 operational spreadsheets examined in field audits, most of them by commercial spreadsheet auditing firms. Collectively, those studies "found errors in 94% of the spreadsheets studied."
These were mostly financial and planning models, not facilities logs, so the exact rates will not carry over. The lesson does: a spreadsheet nobody tests is probably wrong somewhere, and the person who built it is the last to know.
Where spreadsheets and disconnected lists fail in facilities
A bigger risk than a bad formula is a record that never got entered, or got entered in a different list. Public audits of facilities operations keep finding the same pattern, whether the list is a spreadsheet or a separate system nobody reconciles.
Work done but never closed. The Salt Lake County Auditor's October 2025 audit of the county's jail facilities looked at 71 of the longest-open work orders, open between 217 and 499 days. In 40 of them, the work had already been completed by vendors or the county's facilities team, but nobody closed the work order. Maintenance staff sometimes fixed things on the spot "without creating a work order" at all.
Two lists that do not talk. The jail's maintenance team and County Facilities Management used separate work order systems. In a random sample of 31 county work orders for the jail, 29 had no corresponding entry in the jail's system. The auditor's conclusion: "neither system maintained a complete record of jail maintenance activities." The same audit found no parts inventory or tracking system, and 15 of 37 sampled tools on the controlled asset list were listed incorrectly or without enough detail.
Preventive maintenance outside the system. In Washington, D.C., an internal audit of the Department of General Services (RSM, November 2022) found the District "currently utilizes manual processes for tracking, scheduling, and performing preventative maintenance activities, which typically occur outside of the CMMS." The recommendation was to bring them into the maintenance system, starting with manufacturer schedules and warranty requirements for each asset.
The file outgrows the process. In October 2020, Public Health England reported that 15,841 positive COVID-19 cases had been left out of the daily figures because "some files containing positive test results exceeded the maximum file size" of the process that loaded them into central systems. That was a public health data pipeline, not a facilities log, but the failure is the same one a growing spreadsheet invites: nothing warns you when records silently stop making it in.
Signs you have outgrown the spreadsheet
These are our rules of thumb, not a published standard. If three or more are true, the spreadsheet is costing you more than it saves:
- More than one person needs to edit it the same day, and you have already lost changes to someone saving over them.
- You cannot answer "who changed this, and when?" for a closed work order or a PM date.
- The same asset or building appears in more than one file, and they disagree.
- Requests still arrive by phone, text and email, and someone retypes them into the sheet at the end of the day.
- You need to show proof to an auditor, an insurer, a landlord or a client, and the proof is a row anyone could have edited.
- Dates are due before anyone looks. Certificates of insurance, lease notice windows and inspections expire because a spreadsheet cannot remind anyone. See our guides to certificate of insurance tracking and commercial lease critical dates.
How to move off a spreadsheet without losing history
The UK National Archives' migration project checklist (June 2017) was written for records management systems, but its steps fit a facilities move well:
- Decide what to migrate. The checklist starts with "Identify what information is affected" and "Consider whether migration is the right approach." Open work orders, the asset register, PM schedules, vendors and critical dates usually move. Ten years of closed tickets may not need to.
- Clean before you import. Merge duplicate assets, fix names to match the ID list, and make every date a real date. A system will not fix a dirty register; it will just make it searchable.
- Test with real data. The checklist calls for testing "using representative data and data volumes" and checking "that imported data meets requirements." Import one building first and compare counts against the spreadsheet.
- Write down what will not come across. The checklist says to "Document what will be missing from the migration, if anything" and decide whether to keep the old system longer because of it.
- Freeze the old file, do not delete it. Save the final version as read-only and stop editing it on the day you switch, so there is one live record.
Keep the frozen file as long as its records might be needed. For tax purposes, the IRS generally says to keep records for 3 years, employment tax records for at least 4 years, and records relating to property "until the period of limitations expires for the year in which you dispose of the property." For a facilities team, that last rule covers the purchase cost and depreciation records for equipment still in service. Leases, insurance and contracts may need longer, so check with whoever owns those obligations.
If you are deciding what to move to, our comparison of CMMS vs IWMS explains the categories, and work order response time covers the clocks worth tracking once requests are in a system.
How MAFM handles it
When a team starts on MAFM, its buildings, floors, assets, vendors and open requests are imported during configuration, and historical leases can be loaded as documents so the platform extracts the terms. After that, every change carries an audit trail of who changed what and when, and modules can be turned on one at a time instead of on a migration deadline. See the platform overview and the asset and inventory solution.
Frequently asked questions
Can you run facilities management in Excel or Google Sheets?
Yes, for a small team with one editor and a modest number of assets and work orders. Build it with one tab per record type, unique IDs, drop-down lists and real date columns. It starts to break when several people edit at once or when you need an audit trail.
What should a facilities management spreadsheet include?
At minimum: locations, an asset register, work orders, a preventive maintenance schedule, vendors with insurance expiry dates, and lease or contract critical dates. Link the tabs with IDs so every work order points to one asset and one location.
How common are spreadsheet errors?
A Dartmouth study of 50 operational spreadsheets found errors in 0.9% to 1.8% of formula cells, and field audits summarized by Ray Panko found errors in 94% of 85 spreadsheets examined. Most errors are small, but a few in each study were very large.
How long should you keep the old spreadsheet after switching?
Keep a read-only copy at least as long as its records could be needed. The IRS generally points to 3 years for tax records, at least 4 years for employment tax records, and property records until the limitations period ends for the year the property is disposed of.
If your facilities still run on a workbook with a filename like FINAL_v7, take the Field Tour to see the same records in one system, or talk to an operator about what to move first.
