MARKETINGCOACH / FREE WORKING TOOL
25 minutes · Test one change without confusing activity with evidence.
Use a real project. Read each prompt, capture your answer, then choose one next action. No signup required. Your answers are not submitted or saved by this worksheet. Print or save a PDF before leaving; use your browser’s print command if JavaScript is unavailable.
Name one funnel step and one problem, supported by a source and date.
Example: 120 relevant visits produced 6 enquiries last week. Several prospects asked what the service includes.
If we change [one thing] for [audience], then [observable action] may improve because [reason].
Example: Adding scope and price before the booking link may reduce uncertainty. This is a hypothesis, not a forecast.
Record the same-period visits, qualified enquiries, bookings, spend, and data source. Define “qualified” before the test.
Conversion rate = completed actions / eligible opportunities. Keep the denominator consistent.
Name the single change, owner, start/end dates, primary metric, and guardrails such as lead quality, spend or complaints.
Do not change traffic source, pricing, copy and layout together if you want to interpret what happened.
For days 1–7, record opportunities, actions, unusual events and customer comments. Keep small samples visible.
A week is an observation window, not proof of statistical significance. Extend the window if volume is low.
Compare counts and rates, review quality and confounders, then choose keep, revise, stop or gather more evidence. Give the next step an owner and date.
Before/after differences do not establish causation. Record uncertainty alongside the result.