Analytics in plain English

Every platform Meera uses produces numbers about itself — reach, impressions, saves, CTR, follower deltas — and for two years she has done what most one-person brands do: glance at the graphs, feel vaguely good or bad, and change nothing. The problem was never access to data; it's that dashboards answer questions nobody asked. What she needed was an analyst — someone to read the numbers and say what they mean and what to do. As of this course, she has one on retainer. So do you.

The monthly conversation

The core move is simple enough to feel like cheating: export or paste the month's numbers into your assistant — per-post reach and engagement, follower change, profile visits, whatever your platform surfaces — alongside the questions a consultant would charge for:

Here's my Instagram data for March [paste/attach]. My pillars: making,
home, knowledge, shop. This month's calendar leaned on the lamp launch.
1. What performed above and below my baseline, and by how much?
2. Patterns: pillar, format, day, hook style — what's actually driving it?
3. What looks like noise I should ignore?
4. Three concrete changes for April's calendar, with your reasoning.

What returns is the analyst's memo dashboards never write: the making pillar outperforms everything — double it; carousels beat single images 2:1 for you; the launch-week story sequences drove profile visits but the bio link changed too late to catch them; Tuesday's dip is noise, stop reading it. Push back and dig exactly as with any draft — "is the carousel effect just the Ramniwas-ji outlier? recompute without it" — because conversation, not display, is what makes numbers usable by someone whose job isn't numbers. April's calendar (lesson 3) then gets built from the memo, which closes the loop this whole course has been assembling: plan → produce → publish → measure → revise plan. That loop, running monthly, is the entire difference between an account with a strategy and an account with a posting habit.

Three disciplines keep the analyst honest

Verify the arithmetic. Assistants read patterns brilliantly and slip on sums — the companion course explains why in one lesson — so any specific number that will drive a real decision ("engagement fell 34%") gets a ten-second check against the platform's own total before you act on it. Patterns from the model, arithmetic from the source: the everyday-work course's spreadsheet rule, wearing marketing clothes.

Mind what the metrics are for. Reach flatters, likes soothe, but Meera sells lamps: profile visits, link clicks, saves (the "I'll come back for this" signal), DMs, and orders are the chain that pays artisans. Tell your analyst which metrics are goals and which are vanity — literally, in the prompt — or it will cheerfully optimise your feed toward applause. The model inherits your definition of success; give it the honest one.

Respect small numbers. At a few thousand followers, one lucky Reel bends every average, and a model asked for patterns will find them — it's a pattern-finding machine, and it will pattern-match noise with the same fluent confidence as signal (its most famous trait, wearing an analyst's tie). The guard is built into question 3 above — make "what should I ignore?" a standing part of the ask, and treat any single-post insight as a hypothesis for next month's test, not a law.

Do this today: run the monthly conversation on your last 30 days, even mid-month, even with modest numbers. Take the one clearest finding and change one thing in your calendar because of it. The loop starts the first time data actually alters a plan — everything before that was decoration.