Keeping a Source List That Holds Up

Most reporters can tell you who they spoke to for a story the week it runs. Fewer can tell you six months later, with the same confidence, what each person said and when. That gap is where a lot of avoidable trouble starts — not because anyone lied, but because memory quietly rewrites itself under pressure.

A source list sounds like an obvious fix, and it is, but the reason it fails in practice usually has nothing to do with discipline. It fails because reporters build one system for a big investigation and no system at all for the smaller, faster stories that make up most of the week. The habit only holds if it's light enough to survive a deadline.

What a workable record actually needs

It doesn't need to be elaborate. A workable source record needs four things for each conversation: who, when, how they were reached, and what was said that ended up in the piece. That's it. The temptation is to build something closer to a full transcript, and that temptation is usually what kills the habit by the third story.

"The record doesn't need to prove you were right. It needs to prove you were careful."

That distinction matters more than it sounds. A careful record shows your reasoning at the time, with the information you had at the time. Nobody can fault a reporter for a source who turned out to be wrong, if the record shows the claim was checked against what was reasonably available.

Where the habit tends to break

The most common failure point isn't the interview itself — it's the follow-up call, the text exchange, the hallway comment that gets folded into a quote later without a note attached. Those informal moments are exactly the ones worth logging, because they're the ones memory reshapes fastest.

Build the record at the time of contact, not at the time of writing. A single line in a running document, dated, takes fifteen seconds. Reconstructing the same information three weeks later, from memory, takes considerably longer and is considerably less reliable.