Why Every Automation Tester Should Learn Tableau in 2026

import csv, glob
import xml.etree.ElementTree as ET

rows = []
for path in glob.glob("results/build-*/junit.xml"):
    run_id = path.split("/")[1]
    for tc in ET.parse(path).getroot().iter("testcase"):
        if tc.find("failure") is not None or tc.find("error") is not None:
            status = "failed"
        elif tc.find("skipped") is not None:
            status = "skipped"
        else:
            status = "passed"
        rows.append([run_id, tc.get("classname"), tc.get("name"),
                     status, float(tc.get("time", 0))])

with open("test_results.csv", "w", newline="") as f:
    w = csv.writer(f)
    w.writerow(["run_id", "suite", "test_name", "status", "duration_s"])
    w.writerows(rows)

run_id,suite,test_name,status,duration_s
build-1481,checkout,test_apply_coupon,passed,3.92
build-1481,checkout,test_remove_item,failed,6.10
build-1482,checkout,test_apply_coupon,failed,4.21
build-1482,checkout,test_remove_item,passed,5.87

// Pass Rate (Tableau calculated field)
SUM(IF [status] = "passed" THEN 1 ELSE 0 END)
/ SUM(IF [status] <> "skipped" THEN 1 ELSE 0 END)

// Is Failure
IF [status] = "failed" THEN 1 ELSE 0 END

// Flaky: a test that both passed and failed in the last 10 runs
// 1) Filter run_date to the last 10 runs
// 2) Calculated field "Has Pass":
{ FIXED [test_name] : MAX(IF [status] = "passed" THEN 1 ELSE 0 END) }
// 3) Calculated field "Has Fail":
{ FIXED [test_name] : MAX(IF [status] = "failed" THEN 1 ELSE 0 END) }
// 4) Flaky = [Has Pass] = 1 AND [Has Fail] = 1

Release risk dashboard layout (1200 x 800)
+---------------------------------------------------+
| Pass rate | Failed tests | Flaky tests | Avg time |
+--------------------------+------------------------+
| Pass rate trend (line)   | Failures by suite (bar)|
+--------------------------+------------------------+
| Action list: top 10 flaky/failing tests (table)   |
+---------------------------------------------------+

Title: Release readiness — build {latest run_id}
Subtitle: Automated regression, last 30 runs
Tile labels: Pass rate | Failed | Flaky | Avg duration
Footnote: Skipped tests excluded from pass rate.

Test data checklist before connecting Tableau
[ ] One row per test per run
[ ] run_id and run_date present
[ ] status values limited to passed / failed / skipped
[ ] duration in seconds as a number
[ ] suite or module name present
[ ] No personal or customer data in test names

Calculated fields to create
Pass Rate ........ passed / (total - skipped)
Is Failure ....... 1 if failed else 0
Has Pass / Has Fail (FIXED per test)
Flaky ............ Has Pass AND Has Fail
Duration Bucket .. <5s, 5-30s, >30s

Weekly dashboard review questions
1. Did the pass rate go up or down this week, and why?
2. Which three tests were flaky most often?
3. Which suite should get more automation effort?
4. Which slow tests can be split or optimised?
5. Is the data still refreshing correctly?