Reading your analytics correctly
What each chart actually means, and the common ways people misread them.
distro's Analytics tab shows time-series, per-channel breakdowns, and cohort-style retention charts. They're useful if you read them correctly and misleading if you don't.
Posts per week
A simple count. Not a success metric — volume is an input, not an output. Use it to check consistency, not to congratulate yourself for a spike.
Median engagement
The middle post of the week. Less gamed than mean, which gets dragged by a single viral hit. Median is what your typical post looks like, which is what you want to improve.
Virality distribution
A histogram of your posts ranked by virality score (relative to your own median). A healthy shape has a right-tail — a few posts way above median. If your distribution is flat, your posts are all the same shape and nothing's breaking out.
Time-of-day heatmap
48 cells per week (day × hour-slot). Darker = more engagement. Look for concentrationnot just peaks — if only Tuesday 8am is dark, that's a pattern; if every cell is mid-grey, you don't have one yet.
Cross-channel retention
Studio+ only. Rows are cohorts of users who first interacted with a specific post; columns are weeks. Darker = retained. Most useful for debugging whether your top-performing posts convert into durable audience growth or just one-shot vanity spikes.
Common misreads
- Treating median as a ceiling. Median is your baseline. The job is to raise it.
- Optimizing only for the heatmap. Time-of-day is a minor factor compared to hook strength. Post-time can buy you 20%; a better hook buys you 300%.
- Reading weekly data as a trend. 1–4 weeks of data is noise. 12+ weeks is trend. Don't pivot on a single down-week.