Read campaign activity without over-reading the numbers.

Raw opens and unique starters are not the same metric.

A member can open the same verification flow more than once. Raw opens describe total attempts or revisits. Unique starters count a member only once per campaign in the selected window. When you want to understand how many distinct people entered a campaign funnel, unique starters are usually the cleaner denominator.

Compare 30 days with the previous 30 days—not with lifetime totals.

Trend comparisons are most useful when the windows have equal length and do not overlap. A current 30-day period compared with the immediately preceding 30 days gives context for starts, completions, conversion and point spend. A lifetime total belongs in a separate career metric because it answers a different question.

Percentage change also needs context. Going from one starter to two is a 100% increase, but it is still only one additional person. Read the absolute counts beside the percentage instead of treating the percentage as the whole story.

Campaign budget describes capacity inside the exchange.

In a prepaid points exchange, the creator funds a base reward for each eligible completion. Creator-funded spend therefore tracks those base rewards consumed by verified completions. Platform-funded level bonuses or temporary earning boosts should be kept separate so the creator's advertised campaign cost does not silently change.

A benchmark is context, not a promise.

A privacy-conscious benchmark can combine activity from many creators and campaigns, then appear only after minimum sample thresholds are met. That can help answer a broad question such as whether a platform's internal completion flow is running above or below a recent aggregate range.

It cannot guarantee reach, followers, sales, algorithmic distribution, or future performance. Campaign destinations, requested actions, audience fit and external platform conditions differ. When a sample is small, showing no benchmark is more responsible than presenting a fragile number as authoritative.

Change one clear variable when possible.

If starts are healthy but completions fall, inspect the destination and action wording first. If starts are weak, check whether the reward, platform, timing or campaign relevance needs adjustment. If runway is nearly exhausted, decide whether more budget is justified before increasing anything else.

Side-by-side campaign comparison is useful for reviewing naturally different campaigns, but it is not the same thing as a controlled A/B test unless traffic assignment and other variables are actually controlled. Avoid claiming causal certainty from an observational comparison.