Exploratory Session Report
A session report is the written result of one exploratory testing session: the charter, how the time was spent, what was covered, the bugs and open issues found, and what the next session should explore. It is the “reviewable result” of Session-Based Test Management, read at the debriefing [1].
1. Worked example: the session
This constructed session’s charter is explore the coupon system using the Obsessive-Compulsive tour and the Never-and-Always heuristic to discover pricing errors; the time box is 90 minutes. Scope: valid, repeated, combined and expired coupons, with sale prices and cart changes; the shop allows one coupon per order, and the first one applied should stay. The specification’s test cases each applied one coupon; none combined coupons with each other, with sales or with cart changes.
- A valid coupon gives 10% off. The Obsessive-Compulsive tour says repeat it: applied twice, the discount doubles: bug 1.
- Never and Always says a price is never negative, so try the fixed coupon WELCOME5 (5.00 off) on an item on sale at 4.50. The price shows −0.50: bug 2.
- Apply WELCOME5 (fixed 5.00 off), then remove items: the discount stays and the subtotal drops below zero: bug 3.
- In two browser tabs, apply a different coupon in each: both apply: bug 4.
Each test came from the previous result, from the tour or heuristic named in the charter, or from an idea the session prompted.
2. Notes during the session
Raw notes, with times:
09:50 start; set-up: test account, cart, coupon codes · 09:59 testing: SAVE10 ok, 10% off · 10:12 SAVE10 twice → 20%! bug: reproduce, file #101 · 10:18 testing · 10:30 WELCOME5 (5.00 off) on sale item at 4.50 → −0.50, bug #102 · 10:35 testing · 10:48 removed items, discount kept, bug #103 · 10:52 testing · 11:05 two tabs, both coupons applied, bug #104 · 11:08 testing · 11:20 end
The times give the split. Set-up is everything before testing can start: accounts, data, environment. Bug time starts when the triggering test shows the bug and runs to the next testing entry; a bug investigated mid-test counts as bug time, as in SBTM [1]:
- set-up: 09:50–09:59 = 9 min, 9 ÷ 90 = 10%;
- bug investigation and reporting: 6 + 5 + 4 + 3 = 18 min, 18 ÷ 90 = 20%;
- testing: the rest, 90 − 9 − 18 = 63 min, 63 ÷ 90 = 70%.
3. The report
Each field has one rule; the third column is this session’s entry.
| Field | Rule | This session |
|---|---|---|
| Charter | The mission as given, all three parts | Explore the coupon system using the Obsessive-Compulsive tour and the Never-and-Always heuristic to discover pricing errors |
| Tester, date, type | By total session length, set-up and bug time included: normal 60 to 120 minutes, short under 60, long over 120 (this handbook’s convention); a short session is reported with the same fields, debrief outcome and next charter included | A. Tester, 12 March, normal (90 min) |
| Time split | Shares of testing, bug investigation and reporting, and set-up, from the clock times; they sum to 100% | 70%, 20%, 10% |
| Areas covered | What you actually exercised, specifically enough to repeat | Valid and repeated coupons; sale items; removing items after a coupon; two tabs on one cart |
| Bugs | One line each with the bug report number; the bug report holds title, steps, expected and actual result, severity (scale below the table), environment | #101 discount doubles (major); #102 negative price on sale item (critical); #103 discount kept, subtotal below zero (critical); #104 two coupons on one order (major) |
| Issues | Open questions and blockers that are not defects; ideas outside the charter’s scope (inside the scope, they go to Next charter) | May coupons combine with sale prices? Idea: gift cards |
| Next charter | What this charter included but was not reached, all three parts | Explore expired coupons using the Goldilocks heuristic on expiry dates to discover pricing errors |
| Debrief outcome | Filled in after the debriefing | #102 and #103 fixed first; next charter approved; a regression test for each of the four bugs |
Severity: critical, the shop charges an impossible amount (a negative price or total); major, a wrong amount, but the total stays at or above zero (#104: a 40.00 cart with SAVE10 and WELCOME5 costs 31.00 instead of 36.00, too low by one extra discount); minor, a wrong message or display; cosmetic, layout only.
When to use it — and when not
Write one for every chartered session.
Limitations
Areas covered are the tester’s own account, not measured coverage.
How solid is this?
- Where it comes from. Bach’s session sheet [1], adapted.
- What we do not hold. Bach’s original sheet format.
References
- J. Bach,“Session-Based Test Management,” Software Testing and Quality Engineering, vol. 2, no. 6, pp. 32–37, 2000.
Disclaimer: AI is used for text summarization, polishing and explaining. Authors have verified all facts and claims. In case of an error, feel free to file an issue.