The 7-Clue Checklist for Evaluating a Cold Case Lead

Most cold case leads arrive looking like a gift. Someone sends a message, a photograph, or a memory that has been sitting in a drawer for thirty years, and for a moment the whole file seems to rearrange itself around it. The urge is to drop everything and chase it. That urge is usually wrong, not because the lead is false, but because a lead is only worth the time it can survive.

People who work old cases for long enough tend to develop the same habit. Before running anywhere, they ask a fixed set of questions. The questions don't tell you whether a lead is true. They tell you whether it deserves the next week of your life. What follows is the version worth keeping on a single page.

Start With a Filter, Not a Feeling

A lead feels urgent in proportion to how badly you want it to be true. That is the trap, and it catches careful people as often as careless ones. Cold cases attract a particular kind of material: recollections polished smooth by retelling, documents misread by someone hoping they mean something, and every so often a fragment that genuinely changes the shape of the file. A filter lets you tell those apart without becoming cynical about all of them.

Score each of the seven questions below as a clear yes, a partial, or a no. Six or seven yeses earns a folder of its own and a scheduled block of time. Four or five means it goes in the pile you check quarterly. Two or three means you write down exactly what you have, date it, and return to the work you were already doing. The score isn't science. It's a way of noticing which questions you skipped because they were the hard ones.

The Seven Questions

1. Is it specific? Could two people disagree about what the claim says? If not, it's too soft to test.
2. Can someone other than the source check it? Name the route: a record, an object, an independent witness.
3. Where did it come from, and what has happened to it since? First-hand, second-hand, or absorbed from coverage of the case years later?
4. Does it fit the timeline without forcing anything? What else would have to be true if this is accurate?
5. Does it bear on opportunity, or only on character? A person being unpleasant is not a lead. A person being somewhere is.
6. What would disprove it? If nothing could, you're holding a mood rather than a lead.
7. What does it add? New information, independent corroboration, or a repeat of what the file already says?

Specificity: Does the Detail Cost Something?

Look for the detail that costs the person something to say. "There was a car near the bridge that night" is free. It cannot be wrong, which is exactly why it has survived so long in the retelling. "A maroon estate, mismatched front wing, parked facing the wrong way with its lights off" is expensive. It can be checked, and it can be contradicted.

The same rule applies to behaviour. "He seemed nervous" is a description of the narrator's impression. "He asked me twice whether the coast road was still closed" is an event. One of these can be looked up, matched against a weather report or a road closure notice, or put to another witness.

There's a warning attached. Specificity can be manufactured, and confident detail is sometimes borrowed wholesale from old newspaper coverage. A story that is precise in exactly the places the press was precise, and vague everywhere else, is telling you something about its origins. Note where the detail stops.

Verifiability: Where Does the Claim Live?

Ask where the claim physically resides. Most sit in one of three places, and each has its own check.

• Records. Court files, inquest papers, newspaper archives, employment records, planning applications. Ask for the reference, then read the document yourself rather than accepting a summary of it.
• Objects. A letter, a photograph with a date stamp, a vehicle, a piece of clothing. Does it still exist, can it be photographed properly, and who has handled it?
• Memory. The weakest container and the most common. A second, genuinely independent recollection outweighs the same person telling the story better a fourth time.

Write the claim as a single sentence that could be proven false, then list every route to checking it. If you cannot find at least two routes, it isn't a lead yet. It's a story, and stories are welcome, but they belong in a different drawer.

Provenance and Contamination

Ask when they learned it

There is a large difference between someone who saw a thing in 1994 and someone who learned about it in 2011 from a television programme and has since built a memory around it. Neither is lying. Memory is not a recording; it updates, and it borrows. A witness who has read every article about the case will often have absorbed details they now sincerely believe they saw.

Ask how many hands it passed through

Each retelling is a chance for a detail to migrate. By the time a lead reaches you it may have travelled through four people, one of whom was certain about the colour of a coat and three of whom repeated it. Write the account down exactly as you received it, dated, and resist the urge to tidy the phrasing. The odd word is frequently the most useful thing in the message.

Do ask why the information is surfacing now. The answer is often entirely ordinary: a retirement, a death in the family, a house being cleared. Sometimes it is more complicated. Don't assume bad faith, but do write down the answer alongside the claim.

Timeline, Opportunity, and the Boring Scaffolding

Test the lead against the dull material first. Shift patterns, bus and train timetables, weather records, the date a shop closed, the week a road was resurfaced. This paperwork quietly kills more leads than any interview does, and it does so in an afternoon rather than a year.

Then ask the follow-on question: if this is accurate, what else must be true? A claim that requires one person to be in two places at once is not necessarily false, but it now carries a debt, and you should know what the debt is before you build on it. Prefer leads that put somebody somewhere over leads that explain what somebody is like. Character is cheap to assert. Presence at a place and time is expensive, and it is what cases are actually made of.

Falsification: The Lead Should Come With a Way to Break It

The strongest leads you will ever receive arrive with their own destruction built in. "The man at the bus stop" cannot be broken. "The man who bought a ticket at 6:40 that evening" can, and that is precisely what makes it useful.

Ask the source directly, in plain words: what would show that this is wrong? A person who has thought carefully will usually give you something concrete, and often they will have already wondered about it themselves. A person who cannot name a single fact that would change their mind has given you a conviction, not a clue.

Finally, compare the lead against the file. If it duplicates what is already there, that is not automatically useless. Independent duplication is corroboration, and corroboration is worth naming as such. Just be honest in your notes about which one you have.

Before You Spend the Week

Turn the lead into a single sheet before you commit real time to it. The claim in one sentence. The source and the date received. The three checks you intend to run, in the order that costs least first. What would disprove it. The next action, with a time limit attached. A date to revisit it if nothing comes of the first pass.

Then set a stop rule and keep it. "If the archive search produces nothing by the end of next month, this goes back in the pile." Leads multiply; hours do not.

One practical caution. When a lead points towards a living person, anything beyond reading public records can have real consequences for them and for you. Keep the early work to documents and archives, and take proper advice before approaching anyone named. Patience is not a formality here; it is the difference between a careful check and an accusation.

Photo: jarmoluk / Pixabay