The Scenario
You are a growth analyst at a B2B SaaS company. It is Tuesday morning and the head of growth wants a week-over-week trend chart for purchase completions before the 10 AM all-hands.
The data is in Amplitude: 90 days of daily purchase_completed event totals. Your Excel workbook is open but empty. Getting the data in means downloading a CSV from Amplitude, pasting it into the Revenue tab, fixing the headers, and calculating week-over-week change yourself with formulas.
The bad version of Tuesday:
- Open Amplitude, configure the Event Segmentation chart for
purchase_completed, set the date range to 90 days, pick daily interval - Export the CSV
- Open the Excel workbook, navigate to the Revenue tab, paste, delete the Amplitude metadata rows
- Reformat the date column so Excel treats it as dates instead of text
- Add a formula in column C for week-over-week change
- Format column C as a percentage
- The all-hands starts before your chart is finished.
The fast version is one prompt.
The Easy Way: One Prompt in SheetXAI
SheetXAI is an AI agent inside your Excel workbook that connects to Amplitude directly — no CSV export, no paste-and-fix.
Open the SheetXAI sidebar and type:
Fetch event segmentation totals for 'purchase_completed' by day for the past 90 days from Amplitude. Write dates into column A and daily counts into column B of the Revenue tab. In column C, calculate the week-over-week change as a percentage. Flag any week where the change exceeded 15% in either direction with "ALERT" in column D.
SheetXAI calls the Amplitude API, writes the data into the Revenue tab, calculates the derived columns, and flags the anomalies. The chart is data-ready in the time it used to take just to download the CSV.
What You Get
A clean, analysis-ready table in the Revenue tab with four columns:
- Column A — dates in YYYY-MM-DD format, 90 rows
- Column B — daily
purchase_completedtotals from Amplitude - Column C — week-over-week percentage change, calculated inline
- Column D — "ALERT" flag on any week that moved more than 15% in either direction
The flags are grounded in the actual numbers, not a template you filled in. If there is a week with a 40% drop, it shows up before you have to go looking for it.
Need a different event? Tell SheetXAI to swap in checkout_started or any event name in your Amplitude project. The prompt shape is reusable every week.
What If the Data Is Not Quite Ready
Most real growth analyses are messier than 90 days of one event. SheetXAI handles the complexity in the same prompt.
When the event name uses inconsistent casing in Amplitude
Your Amplitude project has both purchase_completed and Purchase_Completed logged by different SDK versions.
Fetch all events from Amplitude where the event name matches 'purchase_completed' or 'Purchase_Completed' (case-insensitive). Combine the daily counts into a single column. Write dates in column A of the Revenue tab and combined daily totals in column B.
When you need the data segmented by country
The head of growth wants to see which geos are driving the trend, not just the global total.
Pull the last 30 days of daily unique users for 'purchase_completed' from Amplitude, segmented by country. Paste the result as a pivot-style table into the Revenue tab with dates in column A and one country per column starting at column B.
When you only want weekdays
The e-commerce pattern is strong on weekdays, noisy on weekends — and the chart looks misleading when weekends are included.
Fetch 90 days of daily 'purchase_completed' totals from Amplitude. Write dates in column A and daily counts in column B of the Revenue tab. Delete or blank out any row where the date falls on a Saturday or Sunday before calculating the week-over-week change in column C.
When the raw data needs cleanup plus a trend calculation plus a summary note all in one pass
You need the trend and a written summary of the most significant week for the exec slide.
Pull 90 days of daily 'purchase_completed' totals from Amplitude into the Revenue tab. Write dates in column A and counts in column B. Calculate week-over-week percentage change in column C. Identify the week with the largest positive change and the week with the largest negative change. Write a two-sentence summary referencing the actual dates and percentages into cell E1.
The pattern: instead of pulling the data first and analyzing it in a second pass, you ask for both in one prompt. SheetXAI handles the API call and the analysis inline.
Try It
Get the 7-day free trial of SheetXAI and open any Excel workbook, then ask it to pull event segmentation data from Amplitude with whatever date range and interval you need. The Amplitude integration is included in every plan. For related workflows, see how to pull funnel conversion data into Excel or the Amplitude in Excel overview.