If frequency has dropped but spend remains the same, it means:
– Reach has increased
– CPM / CPCs have gone up
– Occasionally both
In those scenarios, without considering conversion data, you bump spend up to get back to the desired frequency.
If frequency remains the same, but conversion rate and engagement rates (CTR, VTR, etc.) drop – that’s a sign of ad fatigue and you want to rotate creative and/or adjust targeting.
If sales are far greater than conversions, consider that higher frequency may be over-attributing view-through conversions. This is very common on FB. Consider a retargeting audience where someone will buy every 30 days regardless of advertising. If you have a 1-day view through conversion window, and you have a frequency of 1, you’ll attribute that sale only 1/30th of the time. But if you have a frequency of 30 and show once per day, FB will claim that sale whenever it occurs.
Typically you segment retargeting/prospecting (to segment people who have already shown intent with the brand) and set baseline conversion numbers, then test spend and sales (both attributed and raw) to get a directional sense of lift. If you’ve got bigger budgets, you can run more statistical significance experiments but they would take months at $130/day spend.