Before you start
Answers are aggregate numbers for your store. Skio Data MCP can't list or identify individual customers.
You can only connect to one store at a time.
What makes a good Skio Data MCP prompt?
The best prompts name the exact metrics, date range, and any segment or product you want. Your AI tool decides which analytics question to run, so specific wording leads to accurate answers and vague wording leads to guesses.
Name the metric: "What was my dunning recovery rate over the last 60 days, by failure reason?" works better than "How are payments doing?"
Give a date range: If you don't, the connector defaults to the last 30 days. Each question can cover up to one year.
Use exact names: Refer to segments and products by the name they have in your Skio Dashboard.
Ask follow-ups: You can keep going in the same chat, like "Now break that down by product" or "How does that compare to the month before?"
Use cases
Each use case below includes a ready-to-use prompt. Copy it into your AI tool and adjust the dates, segments, or products to match your store.
Get a weekly performance snapshot
Scenario: You want a quick read on how your subscription business did last week without pulling reports from the dashboard.
Example prompt:
Use Skio to compare the last 7 days with the 7 days before. Include total subscription revenue split into recurring, first-time, and one-time, active subscriptions, new and cancelled subscriptions, churn rate, cancel flow save rate, and dunning recovery rate. Flag anything that changed by more than 10%.Why this works: It names every metric and both date ranges, so your AI tool doesn't have to guess what "how did we do" means. Flagging large changes keeps the answer focused on what needs attention.
Tip: Dunning recovery rate always covers the trailing 6 months, no matter what dates you ask for. Treat it as a trend number, not a weekly one.
Find out why subscribers are cancelling
Scenario: Your churn looks higher than usual and you want to know whether it's a real trend and what's driving it.
Example prompt:
Use Skio to show my churn rate for each of the last 3 months. Then show the top cancellation reasons from my cancel flow for the last 30 days compared with the 30 days before, and tell me which reasons grew the most.Why this works: Churn rate is one number per window, so asking for it month by month gives you a real trend. Pairing it with cancel reasons points you to the cause, which helps you decide which Cancel Flow offers or product changes to try.
Tip: Follow up with "Which save offers are saving the most subscribers?" to see what's already working in your Cancel Flow.
Investigate failed payments
Scenario: You've noticed more subscriptions failing payment and want to know whether it's normal, a customer card issue, or a payment processor problem.
Example prompt:
Use Skio to show how many subscriptions entered dunning each day over the last 30 days, broken down by failure reason. Tell me which failure reasons recover well and which recover poorly, and how many retries it usually takes to recover a payment.Why this works: Breaking failures down by reason separates customer issues (like insufficient funds or expired cards) from setup or processor issues. Knowing how many retries it takes to recover helps you judge whether your Payment Recovery retry schedule needs adjusting.
Tip: If one technical error suddenly dominates, check with your payment processor before changing your retry settings.
Compare products or segments
Scenario: You want to see which products or customer groups perform best so you know where to focus offers and merchandising.
Example prompt:
Use Skio to show subscription revenue, active subscriptions, and cancellations by product for the last 90 days. Then compare my VIP segment with the rest of my store on revenue, average order value, and churn rate for the same period.Why this works: Seeing revenue and cancellations side by side shows products that sell well but don't retain. Segment comparisons tell you whether a group needs different offers or messaging.
Tip: Replace "VIP" with a segment name from your Skio Dashboard. If the name doesn't match exactly, your AI tool may answer for the whole store instead.
Automate reports and combine Skio with other tools
Scenario: You want a recurring report delivered automatically, or you want to combine Skio data with data from other apps you use.
Example prompt:
Every Monday at 8 AM, use Skio to summarize last week's subscription revenue, active subscriptions, cancellations, and failed payments compared with the week before. Post the summary to my #subscriptions Slack channel.Why this works: Skio Data MCP answers questions when asked. Your AI tool's scheduling and other connectors, like Slack, Google Sheets, or Shopify, turn those answers into a report that shows up where your team already works.
Tip: If you also connect Shopify to your AI tool, try "Compare my Skio subscription forecast for the next 30 days with my current Shopify inventory, and flag products that might run out."
More sample prompts
Use these as starting points. Each one maps to analytics Skio Data MCP can answer.
Revenue and forecasts
What's my current MRR, and how much did I add and lose each month over the last 6 months?
How has my subscription revenue trended over the last 90 days, split by recurring, first-time, and one-time revenue?
Based on subscriptions already scheduled to bill, what's my projected revenue for the next 30 days, by product?
Subscribers and retention
How many active subscriptions do I have now compared with January 1, and how many were new vs. cancelled?
At which order number do most subscriptions cancel?
Are subscribers who signed up in the last 3 months retaining better or worse than those who signed up 6 to 9 months ago?
What's my lifetime value by signup cohort for the last 12 months?
Cancel Flow and Winbacks
What percentage of customers who started my cancel flow in the last 30 days were saved, and which offer saved the most?
How is my Winbacks journey performing over the last 90 days, by original cancellation reason?
Products and Build-a-Box
Which products generated the most subscription revenue in the last 30 days?
What's the mix of billing intervals across my subscriptions, by product?
What are the most popular products inside my Build-a-Box subscriptions?
Loyalty and promotions
How much unspent loyalty credit do customers have right now, and how many customers hold a balance?
Do customers who redeem loyalty credit place more orders than customers who don't?
How many active subscriptions are on a volume discount, and what's their average order value?
Common mistakes to avoid
Asking vague questions: A prompt like "How's my business doing?" makes your AI tool pick metrics for you, and it may answer a narrower question than you meant. Name the metrics you care about.
Leaving out the date range: Without one, you get the last 30 days. Always say which period you want, and split anything longer than a year into separate questions.
Expecting every metric to follow your dates: Dunning recovery rate covers the trailing 6 months, Surprise & Delight compares the last 30 days with the prior 30, and most loyalty credit metrics are all-time. Ask your AI tool which date range it used if a number looks off.
Using approximate segment or product names: Names are matched against your store. A close-but-wrong name can return results for the wrong group or the whole store.
Asking for customer-level lists: Skio Data MCP only returns aggregate numbers. To work with individual subscribers, use Skio Dashboard > Tools > Data Export.