Signs the spreadsheet has run its course
- Only one person can answer "when was this last serviced?" without searching.
- There is more than one copy of the register, and nobody is sure which is current.
- Defects are raised by email or phone, and the date they were first reported is hard to prove.
- Contractor reports live in the contractor's portal, not in your records.
- The reason a routine runs at its interval is not written down anywhere.
None of these is a failure of the people involved. They are the limits of a tool that was never built to hold a maintenance history.
The sequence
1. Decide what one record is
Before you move a single row, agree on what an asset is and where it lives. A useful pattern is one record per asset, held at a location you would recognise on a walk: site, building, floor and room. Decide whether a system is one asset or many. A fire hydrant system, for example, can be held as the system with its pumps and valves beneath it. Make these calls once, write them down, and apply them across every site.
2. Clean the register before you move it
Moving a messy register gives you a messy register in a better tool. Work through the sheet and resolve duplicates, assets with no location, and rows that describe a room rather than an asset. For each asset, capture make, model and serial number, and who maintains it, the in house team or a named contractor. Where a detail is not known, mark it as unknown. A blank you can see is better than a guess that looks like a fact.
3. Tie each routine to the rule behind it
This is the step most migrations skip, and it is the one that matters most when someone questions an interval. For each routine, record where its frequency comes from: the manufacturer's recommendations, AS 1851-2012 or an approved performance solution for fire measures in NSW, the public health requirements for warm water systems and cooling towers, or your own risk assessment. A routine that just says "every 6 months" cannot be defended. A routine that names its source can.
4. Bring across history, and label it honestly
Import the history you need to plan the next cycle and to answer the questions you expect: the last service date for each routine, open defects, and recent contractor reports. Keep the original spreadsheets and documents. In the new system, treat imported history as a copy of past records, not as work the system saw happen. That distinction matters later, when a dashboard shows what is up to date. Imported is not the same as verified.
5. Start live work on day one
Do not wait for the history to be perfect. From the day you switch, every new request and every open defect goes into the system, and nowhere else. Live records, created as the work happens, are what build a trustworthy history. A record that shows when a fault was reported, when it was made safe and when it was fixed is worth more than a year of backfilled rows.
6. Bring contractors into the same record
Tell each contractor how jobs will reach them from now on, and where their reports must land. Each report should attach to the job and the asset in your system. Download the reports still sitting in contractor portals and attach them too. Keep your own copy rather than relying on a contractor's portal, because the history has to outlast any contract.
7. Run both for one cycle, then retire the sheet
For one maintenance cycle, check the new system against the old sheet. Every routine that fell due should appear in both. Once they agree, make the spreadsheet read only and archive it with a note saying where the live record now sits. Leaving it editable invites someone to keep using it.
Capture the knowledge that walks out the door
The most valuable thing in a long serving site manager's head is rarely in the spreadsheet. It is the pump that trips if you start it before the valve is open, the roof hatch that needs a key from the tenant, the contractor who must be called before eight. Ask for it while you build the register, and write it as a note on the asset itself, where the next person will find it standing in front of the plant. A history that stays with the asset survives a resignation. One that lives in a person does not.
How long to keep what you move
Different rules set different periods. In NSW, AS 1851 records are kept on site for at least 7 years. A test and tag record is kept at least until the next test. Records for registered plant such as lifts are kept for as long as the plant is used. In aged care, incident records under the Aged Care Rules are kept for 7 years. Across a portfolio, the simplest policy is to keep the full history against each asset for as long as you hold it, and to make sure you can export it if you ever change systems.
A short migration checklist
- One written definition of an asset and its location, applied to every site.
- Duplicates resolved, and every unknown detail marked as unknown.
- Every routine names the rule its frequency comes from.
- Imported history labelled as imported, with the originals kept.
- New requests and open defects go into the system from day one, and nowhere else.
- Contractor reports attach to the job and the asset in your own system.
- The site manager's working knowledge is written as notes on the assets.
- After one cycle, the spreadsheet is read only and archived.