The answer to this question will of course depend heavily on the nature or your organization and its goals, but there are a few universally applicable factors that you should take into account:
The volatility of the data:
If you check your metrics too frequently you’ll end up just reacting to the noise of the data, not the signal. Every website metric is affected by random variability. But, over time, many metrics will point in a certain direction — they’ll contain a signal that tells you something valuable. You want to notice and react to the signal, not the noise. Based on how volatile a particular metric is, think about how much time you’ll likely need to be able to distinguish the signal from the noise. Then don’t check the metric much more frequently than that. For example, if your metric tends to vary randomly quite a bit week-to-week, then you probably won’t want to check on that metric more than once a month.
Your ability to act on insights:
If your organization (or part of your organization) can’t act on analytics insights within a week, then reporting on a weekly basis is likely overkill. You’ll be calling audibles that your teammates won’t able to react to in time. Your reporting schedule should fit your organization’s agility level.
The length of the sales cycle:
If your customer’s typical buying cycle is two weeks, then reporting on a key metric every day is overkill — you won’t have enough data at any one point to tell if a change is really impacting the bottom line. But, if you only report on that key metric once a quarter then you may miss important trends that you could have reacted too had you noticed them earlier.
Seasonality:
If your organization is highly seasonal — most of your conversions happen during certain times of the year — then it may only be important to check those metrics during those times.
The importance of the metrics:
Your most important conversion goals should be checked more frequently than things like mobile usage and traffic by location (unless you’re expecting those metrics to change). If you notice changes in your most important metrics, you may then want to drill down to see if those are due to changes in particular device or location segments, but if a metric isn’t central to your business then don’t bother checking it more than you need to.
