# Case: the first month against the success criteria

> Physics opened on time, but bookings are half the target: separate delivery from result, find where the funnel leaked and write the closing report with three actions.

- Course: [Project Management Without the Bureaucracy: Run One Project from Brief to Closure](https://productovi.com/courses/project-management-without-bureaucracy.md)
- Module: Closing and learning
- Lesson 32/32, ~37 min
- Source: https://productovi.com/courses/project-management-without-bureaucracy/first-month-case

## 1 March: the numbers are in

**Alex Tutor:** Back at Klasna on 1 March 2027: physics has run a full month, operations own it, and today you run the benefits check Leo will read.

**Leo:** We opened on time and within budget. Tell me it worked.

**Sofia:** Delivery worked: opened on 1 February, nine tutors, 586,300 UAH spent against the 600,000 ceiling. The result is another story: 58 packages by 28 February against 120.

**Hanna:** So 13,700 under the ceiling, good. And what share of the target is 58?

**Alex Tutor:** Here's the February funnel. Read it before you explain anything.

**Dataset (CSV):** [Physics, February 2027: the weekly funnel](https://productovi.com/datasets/project-management-without-bureaucracy/first-month-case/physics-february-weekly.csv)

Four calendar weeks after opening. On the physics page a package is bought after the diagnostic test result. A learner starts lessons in the week of purchase, so by 28 February only W1 buyers have finished their fourth week. Plan assumptions from December: 2,500 visits, 40% start the test, 75% finish it, 16% of finishers buy (120 packages).

- `week`: Week of February (W1–W4)
- `from`: First day of the week
- `to`: Last day of the week
- `landing_page_visits`: Visits to the physics landing page
- `tests_started`: Diagnostic tests started
- `tests_completed`: Diagnostic tests completed
- `packages_bought`: Physics packages bought
- `plan_packages`: Packages the plan expected that week (120 ÷ 4)
- `buyers_past_week_four`: Of that week's buyers, how many have finished their fourth week by 28 February
- `buyers_attending_week_four`: Of those, how many still attended in week four

> **Exercise:** What share of the target of 120 packages did physics reach by 28 February? In %, two decimals.

## Where it leaked

**Danylo:** Someone will blame the ads. The visits say otherwise: 2,620 against a plan of 2,500.

**Alex Tutor:** Then follow the funnel step by step against the plan's assumptions: 40% start the test, 75% finish it, 16% of finishers buy. All three were our guesses in December.

**Sofia:** Starts: 1,060 of 2,620, that's 40.46%, on plan. Finishing looks like the gap.

**Alex Tutor:** Compute the completion rate. Then look at it week by week: if it's the same every week, it isn't a launch-week glitch.

**Vira:** It's steady across all four weeks. So it's built into the test, not bad luck.

> **Exercise:** What share of the diagnostic tests started in February was completed? In %, two decimals.

**Alex Tutor:** Now the step after the test: how many finishers bought?

> **Exercise:** Of the families who **completed** the diagnostic test in February, what share bought a package? In %, two decimals.

## Phones, and the step after the test

**Dataset (CSV):** [Diagnostic test by device, February 2027](https://productovi.com/datasets/project-management-without-bureaucracy/first-month-case/physics-test-by-device.csv)

The same month split by the device the test was started on. Totals match the weekly file.

- `device`: Phone or computer
- `tests_started`: Tests started
- `tests_completed`: Tests completed
- `packages_bought`: Packages bought after the test

**Danylo:** Phones are two-thirds of the starts, and only 35% finish there. On computers it's 64.71%.

**Alex Tutor:** That's the biggest leak. Had completion matched the plan's 75% with everything else unchanged, February would have had about 98 packages: 1,060 × 0.75 × 58 ÷ 472.

**Sofia:** And the purchase step: 58 of 472 finishers bought, 12.29% against 16%. With only that step on plan, about 76.

**Vira:** Does the data say why phones drop out? The dropped-connection problem we accepted?

**Alex Tutor:** It says where, not why. The connection problem is one hypothesis; a 20-minute test on a phone is another. Watch five parents take it before you fix anything.

> **Exercise:** Which conclusions does the February data support? Choose all that apply.

**Alex Tutor:** Size the biggest leak: what would 75% completion have given?

> **Exercise:** If test completion had matched the plan's **75%** and every other step stayed as it was in February, about how many packages would have been bought? Round to the nearest whole package.

## Week four: early and small

**Vira:** And 71% still attend in week four. Just under three in four.

**Alex Tutor:** Check how many learners that number covers.

**Sofia:** Only the first week's buyers have finished week four: 15 of 21. The other 37 haven't reached it yet.

**Alex Tutor:** So report it as early: 71% of 21 learners, re-checked on 28 March for all 58. One more learner staying would make it 76%. It's neither a pass nor a fail yet.

**Leo:** Fine. And the tutors? Are they all right?

> **Exercise:** How should the closing report state week-four attendance?

## Tutors with empty slots

**Dataset (CSV):** [Physics tutors, February 2027](https://productovi.com/datasets/project-management-without-bureaucracy/first-month-case/physics-tutors-february.csv)

Nine tutors who passed a demo lesson. The plan assumed about 15 learners each at two lessons a week. Tutors are paid 430 UAH per delivered lesson from the operating budget.

- `tutor_id`: Tutor
- `weekly_slots_offered`: Lesson slots the tutor offered each week
- `learners_assigned`: Learners assigned in February
- `lessons_delivered`: Lessons delivered in February
- `payout_uah`: Payout for February, UAH (430 per lesson)

**Alex Tutor:** Compare the slots the tutors offered every week with the lessons they delivered.

**Danylo:** 240 slots a week, so 960 over the four weeks.

**Vira:** The plan gave each tutor about 15 learners. They got 6.44 on average, and they're paid per delivered lesson. They feel it.

**Alex Tutor:** That's an operations risk now, Roman's, not a project task: tutors who earn little may leave before the spring peak. It goes into the closing report as a handed-over risk with an owner.

> **Exercise:** What share of the lesson slots the tutors offered in February was used for delivered lessons? In %, two decimals.

## The closing report

**Hanna:** Leo will ask whether we should spend more to make the 586,300 pay off.

**Alex Tutor:** What's spent doesn't vote. Decide on what each next step costs and changes from here. More ads into a funnel that leaks at the test buy fewer packages than fixing the test.

**Alex Tutor:** So what goes into the report as the next step?

> **Exercise:** Hanna asks whether Klasna should spend more on advertising in March to make the 586,300 UAH pay off. The February funnel: visits and test starts were on plan; test completion was 44.53 % against the planned 75 % (35 % on phones); the purchase step was 12.29 % against 16 %. Completion at plan alone would have given about 98 packages instead of 58. What do you recommend in the closing report?

**Leo:** Then give me the closing report on one page, with the first three retrospective actions.

**Sofia:** Delivery met, result not met, week four early, where it leaked, what's handed over, three actions with owners and dates.

> **Exercise:** Write the one-page closing report for the physics launch, with the first three retrospective actions.

> **Key points**
>
> - Report the two halves apart: **delivery** met (opened 1 February, 9 tutors, 586,300 of 600,000 UAH), **result** not met (58 of 120 packages, 48.33%).
> - Follow the funnel against the plan's assumptions: traffic and test starts were on plan; the leaks are test completion (44.53% against 75%, phones 35%) and the purchase step (12.29% against 16%).
> - Data says where, not why: watch real parents before fixing.
> - Week-four attendance of 71% covers 21 learners: label it early and re-check on 28 March.
> - Idle tutors (28.33% of slots used) are an operations risk with an owner, not a project task.
> - Money spent doesn't decide the next step; three blameless actions with owners and dates do.

**Alex Tutor:** That closes the project: from a one-page brief in October to a one-page closing report in March. The next launch starts with better templates and real numbers.

- Previous lesson: [The benefits check and a blameless retrospective](https://productovi.com/courses/project-management-without-bureaucracy/benefits-check-and-retrospective.md)

---
Productovi Learning. Free, no sign-up needed to read.