What changed after the promotion?

Capture the starting point

Before a promotion begins, write down its purpose, dates, audience, message and destination. Save the actual copy and image used. A later note saying “social campaign” is too vague to explain why one activity differed from another.

Record the relevant baseline using the same measures you intend to review afterward. If the aim is to bring readers to a sample page, note its ordinary traffic if that information is available. If no reliable baseline exists, say so. Missing information is a limitation to preserve, not a blank to fill with an estimate presented as fact.

Keep observations separate from explanations

An observation might be that visits to the book page rose during a particular week. An explanation might be that an interview caused the increase. The first may be measurable while the second remains uncertain, especially if a newsletter, event and retailer promotion happened at the same time.

Use language that reflects the evidence. “Visits increased during the interview week” is different from “the interview produced every additional visit.” This distinction keeps the record useful when you later compare activities. It also prevents a small coincidence from becoming a rule about what always works.

Note the quality of available measurement. Some systems count repeat visits, some obscure sources and some report with a delay. You do not need to solve every analytical limitation, but you should understand enough to avoid treating unlike numbers as directly comparable.

Record the work around the result

Include actual expenditure, preparation time and any operational problems. A promotion that brings useful attention may still be difficult to repeat if it requires more time than the author can sustain. Conversely, an activity with modest immediate results may produce a useful relationship or reusable asset.

Keep qualitative evidence in its own form. A reader's question, a host's feedback or a request for another event can reveal something that a chart misses. Quote privately received comments only with appropriate permission when using them publicly. In the working record, identify their context so they are not mistaken for a representative survey.

Write a decision, not a verdict

End the record with the next action and the reason for it. You might repeat the activity, change the message, repair a weak destination or stop using a channel that does not fit the book. A useful decision can be modest and conditional.

Review several records together when enough comparable evidence exists. Look for patterns while checking whether the conditions were actually similar. A campaign note is most valuable when it preserves details that memory tends to smooth away: what was tried, what happened, what remains unknown and what you will do differently next time.