L2 · Classify requirements into dimensions and metrics

Lab 2: Classify requirements into dimensions and metrics

Objective

Learn to take a business question and break it into dimensions and metrics.

Why this matters

Business stakeholders ask questions in plain English:

• "How many members completed their profile?"
• "Which search term is most popular?"
• "Are mobile users more likely to submit complaints?"

Your job is to translate these into dimensions and metrics so you can build reports.

Exercise 1: Simple questions

For each question, identify the dimension (what you group by) and metric (what you count).

Question 1: "How many page views did each page get?"

Your answer:

• Dimension: Page Name
• Metric: Page Views

*Why:* You're grouping BY page, counting page views for each.

Question 2: "How many visitors came from organic search?"

Your answer:

• Dimension: Traffic Source (filter to "organic")
• Metric: Visitors

*Why:* You're grouping BY traffic source, counting unique visitors. Filter to "organic" to isolate that segment.

Question 3: "What's the bounce rate by device?"

Your answer:

• Dimension: Device (Mobile, Desktop, Tablet)
• Metric: Bounce Rate

*Why:* You're grouping BY device, looking at bounce rate for each. Bounce rate is calculated per device.

Exercise 2: Example Super questions

Example Super (superannuation fund) wants to understand complaint form submissions. Classify each:

Question A: "How many forms were submitted?"

Your answer:

• Dimension: (None — just a count)
• Metric: Form Submissions (count of event5)

*Why:* No grouping. Just total count.

Question B: "How many form submissions by form type (contact vs complaint)?"

Your answer:

• Dimension: Form Type (or a custom prop/eVar you created for this)
• Metric: Form Submissions

*Why:* Grouping BY form type, counting submissions for each.

Question C: "Are mobile users more likely to complete the form than desktop?"

Your answer:

• Dimension: Device
• Secondary metric: Conversion Rate = Form Submissions / Visitors

*Why:* Grouping BY device. The "likelihood" is a conversion rate (submissions per visitor).

Question D: "Which traffic source brings the most form completions?"

Your answer:

• Dimension: Traffic Source (organic, paid, direct, referral)
• Metric: Form Submissions

*Why:* Grouping BY traffic source, counting form submissions for each.

Exercise 3: Create your own

For each scenario, classify into dimensions and metrics:

Scenario 1: "How much revenue did we make from each campaign?"

Your answer:

• Dimension: Campaign (eVar you set on first hit)
• Metric: Revenue

Scenario 2: "Do members who saw the on-site message have higher engagement?"

Your answer:

• Dimension: Message Exposure (prop/eVar tracking if shown the message)
• Metric: Engagement (could be page views, or a custom event)

Scenario 3: "What's our biggest drop-off point in the form?"

Your answer:

• Dimensions: Form Step (prop tracking which step: intro, personal info, bank details, confirmation)
• Metric: Form Completions (count of events for each step)

*Why:* Compare completion counts at each step to find drop-off.

Exercise 4: Tricky ones

These are harder. Think carefully.

Tricky 1: "How many unique members submitted a form?"

Your answer:

• Dimension: (None)
• Metric: Unique Form Submitters (event5 deduplicated per visitor, not per hit)

*Why:* You're not grouping by anything. You're counting unique visitors who submitted at least once. This is a custom metric.

Tricky 2: "What percentage of search results pages resulted in a click-through?"

Your answer:

• Dimension: (None)
• Metric: Custom: (Clicks on search results) / (Search results page views) * 100

*Why:* This is a conversion rate between two actions. You need to define events for each, then create the calculated metric.

Your measurement task

For Example Super, answer these three questions. Classify each into dimensions and metrics.

1. "How many visits resulted in a form submission?" (Dimension: ?, Metric: ?)
2. "Which site section (Home, Portal, FAQ) gets the most traffic?" (Dimension: ?, Metric: ?)
3. "Do returning members stay longer on the site than new members?" (Dimension: ?, Metric: ?)

Write these down. You'll use them in later labs.

Key insight

Every business question can be translated into dimensions and metrics. The translation is your bridge between business language and analytics data.

---

Next: S3 · Props and eVars