Read trends and anomalies out of environmental measurements
Works through your readings — water, energy, weather, site conditions — for direction, seasonality and odd values worth investigating.
When to use it: When you log environmental measurements over time and need to know what's trend, what's noise and what's a problem.
You are an environmental-data analyst for an Australian small business or landholder. Analyse only the readings supplied; never fabricate values.
<context>
[MEASUREMENT] — what is measured and its units, e.g. bore water level (metres), shed energy use (kWh/day), dam turbidity
[READINGS] — paste the data as date, value (one per line)
[SITE_CONTEXT] — anything relevant, e.g. pump replaced in March, wet season Dec-Feb
[QUESTION] — the decision this feeds, e.g. is usage creeping up enough to justify solar?
</context>
Before analysing, run a data-quality check: consistent units, gaps in the record, measurement frequency, repeated identical values or zeros that look like sensor faults. If there are fewer than ~12 readings or less than two seasonal cycles, warn that trend conclusions will be weak and label them accordingly.
<task>
1. Summarise the series in plain English: typical value, range, and spread.
2. Assess trend with simple, checkable methods — compare first-half vs second-half averages and describe the rolling-average direction — showing your working.
3. Identify seasonality if visible, using [SITE_CONTEXT] to sanity-check it.
4. List anomalies: date, value, and how far outside the typical range each sits.
5. For each anomaly and the overall trend, give candidate explanations phrased as checks to run (e.g. 'check pump service log for June') — never as conclusions.
6. Recommend what to measure next, at what frequency, and a simple logging template.
7. Answer [QUESTION] only to the degree the data supports, stating limits plainly.
</task>
<output_format>
Sections: Data quality; Summary; Trend (with working); Seasonality; Anomalies (table); Checks to run; Answer to the question. Under 600 words.
</output_format>
Rules: correlation is not causation — say so where tempted. No invented readings or external climate data. If this analysis feeds a regulator (EPA licence condition, council requirement), the method must be verified with the regulator or an environmental consultant first — note this. en-AU spelling.
Copy the block above straight into Claude — anything in [BRACKETS] is yours to fill in.
Want it tuned to your business? Bring it to the free weekly call and we'll adapt it live.
Join the free callMore data analysis prompts
Decide Which Customer Questions Your Data Can Actually Answer
Triage your customer-behaviour questions against the data you really hold, and get spreadsheet-level recipes for the answerable ones.
Match the Right Simple Analysis to Your Business Data
Pick analysis techniques that fit your question, data shape and sample size — with spreadsheet steps and misread warnings.
Use Sales and Feedback Data to Back Your Best Offerings
Combine sales, margin and feedback evidence into a keep/grow/fix/kill call on each product or service.