Maps and records before the person who knows them retires

You know the moment: a main breaks at 2 AM in a part of town you barely know. Water is running down the street, and you are reading a map by flashlight. The person who could name every valve from memory is asleep, retired, or gone.

When the map cannot answer fast, you do not just lose a few minutes. You shut down a bigger area than the break required, and one main break turns into an all-night outage with a phone that will not stop ringing.

The fix is a minimum viable record of your system, built now, before the next retirement carries the knowledge out the door with it.

What do you capture first?

Start with three things: your mains, your valves and hydrants, and a written record of every leak repair. Together those three make up what the source material calls the minimum viable map, a skeleton you build once and then improve through normal work.

• Mains: the major routes, plus size and material where you already know them.
• Valves and hydrants: location, an ID you can search on later, and current status.
• Leak repairs: what happened and what you did, even if the note is one line.

Service lines usually come later, once the mains and valves are solid. The source material treats service lines as the next layer of the same map, not the starting point, so do not let them hold up the first pass.

What goes on each record?

Every asset or event record needs the same five things. Call it the minimum viable record:

• When and where it happened, marked on the map.
• What asset was involved, tied to an Asset ID.
• What happened and the likely reason, with "Unknown" as a fair answer when you are not sure.
• What you did, plus any follow-up still needed.
• A photo or document.

Asset ID is the glue. It ties a break, a photo, and a repair back to the same piece of pipe or the same valve, so later you can pull every record for that segment in one search instead of digging through file cabinets.

Unknown is allowed as an answer. Guessing is not, because a guess gets written down, then read later as if it were fact, by someone who was not there.

For an isolation valve, capture the Asset ID, valve type, size, status, and where the location came from (field GPS, an as-built, or someone's memory). If it is a valve you would use to isolate a break, add which way it normally sits, which direction it closes, how many turns, and whether it turned easy, moderate, hard, or was seized the last time anyone tried it.

For a hydrant, capture the Asset ID, type, status, location source, whether it operates, its condition, and whether it drains. For a main, capture the Asset ID, diameter, material, status, location source, and pressure zone, marked Unknown rather than guessed.

The quirks matter as much as the basics. A valve buried under gravel, a hydrant that needs a special wrench, a main that shares a trench with someone else's line: these are the details that otherwise live only in one crew member's head. The source material treats them as part of the same record, not an afterthought, with a field for box or pit condition, a field for access notes, and a photo for exactly this reason.

How do you capture as you work?

Capture happens in the field, while the work is fresh, not back at the office trying to remember it later. For every break, the source material teaches one habit: record the Five W's and How.

• Who worked it: your crew or an outside contractor.
• What the issue was, and what you did about it.
• When it happened: date and time.
• Where it happened, precise enough to find the pipe segment later.
• Why it likely happened, or "Unknown" if you are not sure.
• How you fixed it: materials used, how long the water was off, and a cost estimate if you can make one.

Two habits keep this fast enough to stick: pick from a preset list rather than typing a fresh phrase for anything that repeats, material, status, cause, and hold each record to two minutes or less. Ten different spellings of "ductile iron" add up to a database nobody can search. A short note covers whatever does not fit a preset list.

How do you check an old map against the ground?

Walk or drive the area with the paper map in hand. Ask the same four questions an operator asks on a call: where am I, what pressure zone am I in, what facility feeds this area, and which valves would isolate a break here.

If you cannot find a valve on the map that you know is on the ground, or you find one on the ground that is not on the map, that mismatch is itself a data point. Write it down instead of working around it.

For install dates and history nobody wrote down, two low-tech moves fill in a lot of the picture. Most systems were built a subdivision at a time, so county records give a rough install-year map. A short crew meeting, asking who remembers any breaks on a given side of town in the last five years, surfaces incidents that were never logged anywhere.

What keeps a record alive after someone retires?

Two things keep a record alive: it has to exist somewhere other than one person's memory, and it has to be handed over on purpose instead of assumed. The source material points to two real losses that make the point plainly: one small system's maps and customer records burned in a fire and were never recreated, and another lost its only operator with no passwords or accounting records left behind.

That is why record ownership belongs in writing, not in good intentions. When a system changes hands, spell out in the contract or handoff agreement exactly which records the outgoing operator must turn over, and require paper records to come in with each invoice rather than as a promise for later. A common experience on takeover is an outdated paper map with mains and valves sketched in, plus a meter-read sheet, and nothing else.

Some operators worry that writing down what they know makes them replaceable. It is the opposite: undocumented systems produce burnout and a steady stream of after-hours calls, and every retirement without a handoff plan leaves a knowledge cliff for whoever is left. What actually protects your position is a system that runs well with you in it, not a map that only exists between your ears.

Why not wait for a perfect map?

Because usable now beats accurate someday. The source material is direct about it: do not let perfect stop useful.

A photo plus two tape measurements is often good enough for a field decision, even though it falls short of survey-grade GPS. Finding bad data the first time you check the map against the ground is not a failure. It is the system doing its job, telling you exactly what to go fix next.

This works whichever tool your system uses. A laminated map with a valve log in a three-ring binder can hold the same minimum viable record as a spreadsheet or a GIS layer, because the fields matter more than the software. Move to a fancier tool once the habit of filling in those fields is already running, not before.

Once your mains, valves, and leak history are written down in a form someone besides you can search, the next move is turning that map into a working exercise and testing program. See valve and hydrant programs for how to run it.