Where the first budget goes

A small museum, a parish archive, a historic house with a collection and no conservator. There is some money, not much, and a general sense that something should be done. The question is what to buy, and the honest answer is not the thing most people buy first.

The order, and the reasoning

The ranking below follows from two properties of each agent: how fast it acts, and whether its damage can be undone. Fast and irreversible comes first. Slow and expensive to fix comes later, on the grounds that a year of measurement will tell you whether it needs fixing at all.

Nothing. Walk the building first. Before any purchase, spend a day with a notebook looking at gutters, roof valleys, pipe runs above stores, the underside of every shelf against an external wall, and anything stored on the floor. Water damage is the fastest catastrophic loss available and most of it is preventable by a builder rather than by an instrument. This step costs nothing and regularly finds the largest single risk in the building.

Traps. A set of numbered blunder traps, a plan of the building, and a register. This is the cheapest monitoring in the field and it detects the fastest irreversible damage there is. A moth population found in month two is a treatment; found in year three it is a loss. Insect monitoring traps covers deployment.

A data logger, or two, plus one outdoors. Not a climate control system. A logger, in the store that matters most, plus one outside so the indoor record is interpretable. This buys the information that decides every subsequent purchase, and buying it first means the later decisions are evidence-based rather than intuitive. Data loggers covers the specification.

Boxes. Enclosures for whatever is currently on open shelving. A box buffers humidity, excludes light entirely, blocks dust and insects, and means the object is handled by its container rather than directly. Nothing else in this list does four jobs at once. Acid-free boxes and tissue covers what to specify and, importantly, when buffered board is wrong.

Blinds, or UV film, if anything is on display near a window. Light damage is cumulative and permanent, and daylight is nearly always the dominant source. This is cheap and it is the one intervention whose benefit begins immediately. UV filter film covers verification, which matters because an unmeasured film is an act of faith.

A light meter, if there is a display programme at all. Only once there is something to measure. A meter without a display policy produces numbers nobody acts on.

What comes later, on purpose

Climate equipment. A dehumidifier or humidifier bought before a year of logging is a guess, and quite often an expensive guess against a problem the building does not have. When the record shows a genuine sustained problem, the equipment can be sized to the measured load rather than to the room volume, which is where most purchases go wrong. Dehumidifiers covers the sizing question.

Cases. Display cases are expensive and easy to specify badly. A case built from the wrong material and sealed well is worse than no case at all, because it concentrates its own emissions around the object. Do this after the Oddy test conversation, not before.

Wireless monitoring. Worth it when there is active equipment whose failure needs to be known about the same day. Before there is any equipment to fail, it is a subscription for a convenience.

The pattern

Every item near the top of this list produces information or removes an immediate risk, and costs very little. Every item near the bottom is a piece of equipment that only makes sense once the information exists.

That is not a coincidence, and it is the whole argument. The most common expensive error in small collections is buying a machine to solve a problem nobody has measured. The second most common is spending everything on the one agent that has an obvious product attached to it, while the agent with no product attached quietly does the damage.

A year of logging, a set of traps and a store full of boxes is an unglamorous programme. It is also, for most collections, the one that would have prevented most of what has actually been lost.