Create a baseline before changing a business workflow
AI-assisted editorial field note · Educational planning framework
What should an experiment baseline record?
Record the current outcome under the conditions relevant to the experiment: the task, period, eligible cases and important constraints. A baseline makes the comparison more interpretable, but it does not automatically isolate the effect of your change. Preserve the definition and note unusual events so later improvement is not attributed to a test when the environment also changed.
Put it into practice
Choose comparable work
Use cases resembling the intended test audience or workflow. Combining unrelated services can hide differences in effort, timing and completion requirements.
Define measurement
Specify the start, end and exclusions. For elapsed time, distinguish active work from waiting if the difference affects the decision.
Capture context
Note staffing, demand, input quality and exceptional events. These conditions can alter the outcome even when the process itself remains unchanged.
Preserve the records
Keep the original observation period and calculation available. Avoid recreating the baseline afterward from selective memory or the most favourable comparison week.
A worked example
Illustrative scenario — not a customer case study
A team measures approval waiting across recent comparable reports before testing named reviewers. The trial coincides with a quieter week, so the closeout notes lower volume as another possible influence. The baseline helps frame the change but does not justify a confident causal claim on its own.
Make the next decision
Start the comparison when the baseline definition is stable and the test can observe similar work. Use uncertainty notes or a better-designed follow-up when conditions differ materially.
A mistake to avoid
Do not choose the worst historical week only because it creates a larger apparent improvement. Explain the comparison period and why it is relevant.
What if no historical data exists?
Collect a short prospective baseline or document the missing evidence. A later test can still be useful, but its conclusion should reflect the weaker comparison.
Take the next step
Check the broader business context with the free six-question assessment. Use your own evidence to choose the next action.
Have a correction? Email admin@acceleratelink.com with this page's URL and the detail.