| # Case Study 1 — Acquisition Analysis |
|
|
| **Estimated time:** 30 min |
|
|
| ## Background |
|
|
| CrowdSec Console is the web interface used by security teams to manage their CrowdSec deployments. Organizations register with a plan tier and accumulate daily activity on the platform. Understanding registration volume, engagement, and upgrade conversion is key to guiding growth decisions. |
|
|
| ## Dataset |
|
|
| **File:** `../datasets/console_users_acquisition.csv.gz` |
|
|
| This is a **daily time-series** dataset: each row represents one organization on one calendar day, covering **2025-01-01 → 2025-06-01** (~22 weeks, 16 146 unique organizations). |
|
|
| ### Field reference |
|
|
| | Column | Type | Description | |
| |---|---|---| |
| | `Date` | date | Calendar day of the observation (`YYYY-MM-DD`) | |
| | `Organization ID` | string (UUID) | Unique identifier for the organization | |
| | `Plan Type` | categorical | Subscription tier on that day: `COMMUNITY` (free, 99%), `SECOPS` (paid mid-tier), `ENTERPRISE` (paid top-tier) | |
| | `Industry Sector` | categorical | Self-reported industry (14 categories + ~2.4% missing). Examples: `PERSONAL_USE`, `IT_AND_SERVICES`, `HOSTING`, `MSSP`, `EDUCATION`, … | |
| | `Org Created At` | timestamp (ISO8601) | When the organization first registered — use this as the **acquisition date** | |
| | `N Signals` | integer | Security signals generated by this org on this day | |
| | `N Active Engines` | integer | CrowdSec engines actively sending telemetry on this day | |
| | `N Attached Engines` | integer | Engines registered (attached) to the org, active or not | |
| | `N Cti Queries` | integer | CTI (Cyber Threat Intelligence) API queries made on this day | |
| | `N Users Logged In` | integer | Number of users from this org who logged in on this day | |
|
|
| > **Important:** The dataset covers a **6-month observation window** (2025-01-01 → 2025-06-01) and contains **only organizations acquired during this period** — i.e., `Org Created At` falls within the window for every org. There is no historical backfill of older accounts. |
| > |
| > **Note:** `Plan Type` can change over time for the same org (e.g., COMMUNITY → SECOPS after an upgrade). There are 179 such transitions in the dataset. |
|
|
| --- |
|
|
| ## Challenges |
|
|
| Analyze the dataset in order to extract key information that would help monitoring core marketing and product efficiency in terms of: |
| * acquisition : the number of new accounts created in number and quality |
| * engagement : how much the console use the features |
| * conversion : number of paying users |
| * churn : users leaving the console |
|
|
| --- |
|
|
| ## Expected Output |
|
|
| - A document with analysis mixing charts, data and text. Be creative! You can use the tool of your choice (google doc, jupyter notebook, ... as you see fit) |
|
|
