Skip to main content

Workflows

Start with a question your team asks every week.

Choose an investigation, check the data it needs, and take the prompt to your preferred AI tool.

Browse prompts Build on the foundation

Weekly trading review

Bring revenue, orders, and acquisition costs into the same review. Check the definitions before deciding what needs attention.

What to bring to the review

  • Week-over-week revenue, orders, and acquisition costs
  • Definitions, source coverage, and reporting cutoff
  • Follow-up questions for the next team review

Start with this prompt

Prepare a trading review for the last complete week. Compare revenue, orders, and acquisition costs with the previous week. Use SourceMedium definitions, state the stores and timezone, and flag missing data before drawing conclusions.

Suggested prompt for connected SourceMedium data.

Data requirements and checks

Data you need

Order history and connected advertising spend. Agree the stores, reporting currency, and weekly cutoff.

Inspect the definitions

Check before you act

  • Use complete, comparable reporting periods.
  • Inspect the metric definitions and source coverage.
  • Separate observed changes from explanations that need investigation.

Channel performance review

Compare spend and customer acquisition across channels, then identify where to investigate next.

What to bring to the review

  • Spend and new-customer changes by channel
  • Attribution method and coverage gaps
  • Questions to resolve before changing spend

Start with this prompt

Compare acquisition performance for the last complete week against the week before. Look up the available definitions and data first. Show spend and new-customer measures by supported channel, identify coverage gaps, and distinguish attribution from causal lift.

Suggested prompt for connected SourceMedium data.

Data requirements and checks

Data you need

Connected ad accounts, order history, and the identity signals required by your selected attribution model.

Inspect the definitions

Check before you act

  • Confirm which ad accounts and stores are included.
  • Use the same attribution method and reporting dates.
  • Attribution assigns credit. To learn whether an ad caused more sales, use an experiment or a separate causal analysis.

Cohort retention review

Compare repeat purchase behavior at the same cohort age so newer customers are not mistaken for weaker customers.

What to bring to the review

  • Repeat purchase comparisons at equal cohort ages
  • Observation windows and incomplete history
  • Customer segments that merit investigation

Start with this prompt

Compare repeat purchase behavior across first-purchase cohorts using the SourceMedium definitions available to this account. Compare cohorts at the same age, state the observation window, and flag incomplete history. Show the evidence behind the comparison.

Suggested prompt for connected SourceMedium data.

Data requirements and checks

Data you need

Order and customer history, an agreed first-purchase definition, and enough history for the comparison period.

Inspect the definitions

Check before you act

  • Compare cohorts at equal ages.
  • Account for incomplete historical coverage.
  • Confirm the valid-order and first-purchase definitions.

Take a prompt to your AI.

Connect the assistant you already use, or ask SourceMedium Agent in Slack.

Connect your AI Explore the Slack analyst

Build on the foundation. Keep your work.

Pro adds direct BigQuery access so your team can query modeled tables, connect compatible tools, and build on the documented schema. Custom extensions are scoped separately. Agree ownership and validation before adding a new source.

Explore the data foundationUse coding-agent skills

A customer's build-versus-buy decision

Wild Earth's team wanted reliable cohorting and drill-down without maintaining every layer of its data stack.

Read the Wild Earth story

Go deeper on attribution

Stop debating which channel drove the sale.

Your team has more attribution data than ever: platform reporting, GA4, third-party tools, and post-purchase surveys. SourceMedium helps you compare those signals using documented logic you can inspect.

Read the use case

Ready to stop debating the numbers?

In 30 minutes, we will

  1. Map your current systems and reporting gaps.

  2. Show one workflow relevant to your team.

  3. Outline likely fit, implementation scope, and next steps.

Book a walkthrough

Share a little context, then choose a time immediately after submitting.

Takes under a minute. Then pick a time.

How we use your details