The Scenario
It's Monday at 9:15 AM and you've just joined the weekly marketing stand-up. Someone pulls up the leadership dashboard in Databox and the campaign numbers are still from last week. All eyes drift to you — you own the campaign tracking workbook.
The data is in the workbook. It has been in the workbook since Friday. The problem is getting it from column E of the CampaignWeekly worksheet into a Databox dataset, formatted exactly the way Databox expects, before this meeting gets any longer.
The bad version:
- Open Databox, navigate to Custom Data Sources, find the right dataset, and manually enter or paste the rows one by one — realizing after the third row that the date format is wrong.
- Go back to the workbook, reformat the date column from MM/DD/YYYY to ISO 8601, copy the rows again, and attempt the import a second time.
- Discover that the "clicks" column contains some cells formatted as text (not numbers), which fails the upload validator, and spend the next ten minutes hunting down which rows are the culprits.
The meeting is still going. The dashboard is still stale. And you're doing data janitorial work that has nothing to do with the analysis you were hired to do.
The Easy Way: One Prompt in SheetXAI
SheetXAI is an AI agent that lives inside your Excel workbook. It reads your data and talks to Databox for you — formatting the rows, handling the primary key, and pushing the batch without you ever leaving the workbook.
Open the SheetXAI sidebar and paste this:
Push all rows in the CampaignWeekly sheet to my Databox dataset ID 'ds_campaign_weekly', using the value in column A (week) as the primary key and columns B through E (impressions, clicks, conversions, spend) as metrics.
SheetXAI reads the worksheet, validates the schema, formats each row as a Databox data point, and sends the batch. Any rows with formatting issues are flagged in column F before the push runs, so nothing lands in Databox half-baked.
What You Get
- A completed push to the Databox dataset, confirmed with a row count in the sidebar.
- Column F populated with status labels: "pushed" for successful rows, "skipped — type mismatch" for any that need attention.
- The Databox dashboard reflects the new data as soon as the push completes — no manual refresh required.
- A summary of which metrics were sent, so you can verify the right columns landed in the right fields.
What If the Data Is Not Quite Ready
The date column is in mixed formats
Some rows use MM/DD/YYYY, others are plain text like "Week of April 14th." Databox rejects both formats.
Normalize the date column in column A of the CampaignWeekly sheet to ISO 8601 format (YYYY-MM-DD), then push all rows to Databox dataset 'ds_campaign_weekly' using the normalized date as the primary key.
The clicks column contains text-formatted numbers
Excel-style number formatting artifacts crept in from a paste. Databox's Push API treats them as strings.
In the CampaignWeekly sheet, convert columns B through E from text to numeric values where they aren't already, then push all rows to Databox dataset 'ds_campaign_weekly' with column A as the primary key.
The data lives across two worksheets — weekly summary and a channel breakdown
You need to join them before the push so each row in Databox carries both summary metrics and channel attribution.
Join the WeeklySummary sheet and the ChannelBreakdown sheet from this workbook on the week column (column A in both), write the joined result to a new sheet called MergedData, then push all rows from MergedData to Databox dataset 'ds_campaign_full'.
The workbook has rows for multiple campaigns and you only want one
Databox dataset 'ds_campaign_weekly' should only receive rows where column F says "Paid Search," not all campaigns.
Filter the CampaignWeekly sheet to only rows where column F equals "Paid Search", normalize dates in column A to ISO 8601, convert columns B through E to numeric, then push the filtered rows to Databox dataset 'ds_campaign_weekly' using the normalized date as the primary key.
One prompt, one workflow — the filter, the cleanup, and the push happen together so you're not running three separate operations and hoping they stay in sync.
Try It
Get the 7-day free trial of SheetXAI and open your campaign tracking workbook, then ask it to push this week's rows to your Databox dataset. You can also look at how to load an MRR breakdown into Databox or see the full Databox integration overview.