Tines Academy › Lesson
Add filters and detail
Add filters and let people open the responses behind the charts, then check that the dashboard shows the right results.
Objective: Define which sections each filter affects and check that filtered results and response lists match the data.
Let people narrow it down
Maya’s team leads have follow-up questions. Is the percentage of negative ratings for “Shipping” rising across every channel, or only on “chat?” Which responses make up that tall “Billing” bar? Filters let them focus on a product area or channel; selecting a bar lets them read the responses behind it.
Decide which sections each filter affects, and make that clear on the page. On Maya’s dashboard, product area and channel filter everything except the data-quality totals.
The period filter changes only the weekly chart and product area bars. The headline number still compares the last full week with the week before, and the response table still covers the last seven days.
Keep the selected filters visible so people can tell what they’re looking at and clear them when they want the full picture again.
Add the filters and detail
Choose the tab that matches your setup, then send the prompt in workflow chat.
Add filters for product area, channel, and period: the last 4, 8, or 12 full weeks.
Product area and channel should filter the headline number, charts, and response lists, but not the data-quality totals.
The period filter should affect only the weekly chart, product area bars, and the responses opened from a bar. Label it so this is clear. Keep the headline number's weekly comparison and the follow-up table's last-seven-days period unchanged.
Start with all product areas, all channels, and 12 full weeks selected. Show the active filters and provide a way to clear them.
When someone selects a bar, open a separate list of every negative response it counts, using the same filters and dates. Don't replace the seven-day follow-up table. Let people open the matching ticket using its real link. Show full comments only where they were approved and loaded.
When a section has no matching responses in its period, explain that rather than showing an empty chart without a message. Don't show a percentage as 0% when there are no responses to calculate it from.
Test filter combinations, period changes, bar details, clearing filters, and the no-matching-responses message where possible. Compare the results with the prepared data and check that the data-quality totals stay unchanged.
Report what you tested, what passed or failed, and anything you couldn't verify. Don't change saved data to create test cases. Keep the preparation rules, access restrictions, and rest of the dashboard unchanged. Leave the workflow as a draft.
Add filters for product area, channel, and period: the last 4, 8, or 12 full weeks.
Product area and channel should filter the headline number, charts, and response lists, but not the data-quality totals.
The period filter should affect only the weekly chart, product area bars, and the responses opened from a bar. Label it so this is clear. Keep the headline number's weekly comparison and the follow-up table's last-seven-days period unchanged.
Start with all product areas, all channels, and 12 full weeks selected. Show the active filters and provide a way to clear them.
When someone selects a bar, open a separate list of every negative response it counts, using the same filters and dates. Don't replace the seven-day follow-up table. Let people open each fictional response and read its full comment. Don't invent ticket links. Keep the "Test data" label visible.
When a section has no matching responses in its period, explain that rather than showing an empty chart without a message. Don't show a percentage as 0% when there are no responses to calculate it from.
Test filter combinations, period changes, bar details, clearing filters, and the no-matching-responses message where possible. Compare the results with the prepared data and check that the data-quality totals stay unchanged.
Report what you tested, what passed or failed, and anything you couldn't verify. Don't change saved data to create test cases. Keep the preparation rules, access restrictions, and rest of the dashboard unchanged. Leave the workflow as a draft.
Test the filters
Read the build summary and test results in workflow chat. Check which sections each filter affects and note anything Tines 3B couldn’t verify.
In Browser, try the checks below for Maya’s data. Use your own product areas and channels for your data.
Narrow the results
Choose “Billing” and “chat,” or another combination with responses in your prepared data.
The headline number, charts, and table should show only responses matching both selections. The data-quality totals should stay unchanged.
Compare one displayed count with the matching prepared responses.
Change the period
Select four full weeks.
The weekly chart and product area bars should cover that period. The headline number should still compare the last full week with the week before, and the follow-up table should still cover the last seven days.
Open the responses behind a bar
With a channel filter still selected, open a bar’s response list.
Check that:
The number of responses matches the bar’s count.
Every response matches the selected product area, channel, and dates.
The separate seven-day follow-up table keeps its original period.
Open one response and check that it matches the filters.
Check an empty result
Try a product area and channel combination with no matching prepared responses.
Each section with no matches should explain that. The data-quality totals should stay unchanged.
A section can be empty while another has results because they cover different periods. If you can’t find an empty example, leave this check marked as unverified.
Clear the filters
Clear the selections.
All product areas, all channels, and 12 full weeks should be selected again. Any open bar details should close or update to match, rather than keep showing the old selection.
If something doesn’t match
Tell Tines 3B which filters you selected, what you expected, and what you saw. Ask for a fix that keeps the other filters, preparation rules, and access settings unchanged.
Repeat the affected check and ask it to record unfinished checks in Readme.
What to do if something still feels fuzzy
You added filters so Maya’s team can focus on a product area, channel, or period and open the responses behind a bar. You also tried the filters and checked which sections changed. Next, you’ll add scheduled updates and make sure the dashboard warns people when its data is out of date.
Build it safely: You compared the filtered results with what you expected in the Browser tab, rather than relying on the build summary alone.