# An A/B test end to end: the Helion readout — Hypothesis Testing Lab

## About these files

Helion Energy's Portal v2 experiment: 12,000 customers randomised between the old billing portal and the new one, the call-centre agents before and after, and the contact-channel mix. Generated by Charles Sambo's enterprise notebook from seed 20260821.

Synthetic data: no real customer, policy or claim is in them, and the companies are fictional.

## `helion-portal-v2-customers.csv`

One row per enrolled customer, randomised between the old portal (control) and Portal v2 (treatment) for one billing cycle. Rows: 12,000.

| Column | What it holds |
|---|---|
| `customer_id` | The customer, 1 to 12,000. |
| `region` | North, South, East or West. |
| `segment` | Residential, SME or Corporate. |
| `variant` | The arm: control (the old portal) or treatment (Portal v2). |
| `went_paperless` | 1 if the customer switched to paperless billing during the cycle, 0 if not. |
| `days_to_payment` | Days from the bill to its payment, two decimals. |
| `payment_method` | Direct Debit, Card, Bank Transfer or Cash. |

## `helion-agent-calls.csv`

One row per call-centre agent, measured in the four weeks before the rollout and the four weeks after. Rows: 80.

| Column | What it holds |
|---|---|
| `agent_id` | The agent, 1 to 80. |
| `calls_pre` | Average daily calls before the rollout, one decimal. |
| `calls_post` | Average daily calls after the rollout, one decimal. |

## `helion-channel-mix.csv`

One row per contact channel over the experiment window, with the share the capacity plan forecast. Rows: 4.

| Column | What it holds |
|---|---|
| `channel` | Phone, Chat, Email or Self-serve. |
| `observed` | Contacts through the channel in the window. |
| `forecast_share` | The capacity plan's share for the channel. |

Every number in the chapter was computed from these files by the lab's engine, and checked against scipy 1.11.4 and statsmodels 0.14.0.

The chapter: https://hypothesis-testing-lab.azurewebsites.net/learn/ab-test-end-to-end