KPI & Metrics — Interview Questions¶
KPI questions test whether you think like an analyst or like an order-taker. The best answers define the metric precisely, explain what drives it, identify its limitations, and suggest complementary metrics.
Tags: #KPI #Metrics #BusinessAnalysis #InterviewPrep
Foundations¶
[Beginner] What is the difference between a KPI and a metric?
Show answer
- A metric is any quantitative measurement: page views, order count, support tickets.
- A KPI (Key Performance Indicator) is a metric tied to a specific business objective, with a target and a decision attached to it.
Every KPI is a metric. Not every metric is a KPI.
Example: - Metric: "Number of support tickets" — just a count - KPI: "First Contact Resolution Rate ≥ 85%" — tied to the goal of resolving issues quickly, with a threshold that triggers action if not met
Interview signal: candidates who confuse KPIs and metrics tend to build dashboards full of vanity metrics. Show you understand that a KPI must be actionable: seeing it turn red should tell you exactly what to do.
[Beginner] What makes a good KPI? What is the SMART framework?
Show answer
A good KPI is:
| Letter | Meaning | Applied to "Increase revenue" |
|---|---|---|
| Specific | Precisely defined | Revenue from new customers (not all revenue) |
| Measurable | Can be quantified | In £, tracked weekly |
| Achievable | Realistic target | +15% (not +150%) |
| Relevant | Linked to business strategy | Supports the goal of expanding the customer base |
| Time-bound | Has a deadline | By end of Q2 2024 |
Interview extension: SMART is a starting point. Also ask: is it controllable? (Can the team actually influence it?) Is it leading or lagging?
[Mid-level] What is the difference between a leading and lagging indicator?
Show answer
- Lagging indicator: measures what already happened. Shows outcome. Can't be influenced once measured.
- Leading indicator: predicts what will happen. Shows inputs or early signals. Can be influenced now.
| Lagging | Leading |
|---|---|
| Monthly revenue | Pipeline value, number of qualified leads |
| Customer churn rate | NPS score, product usage frequency, support ticket volume |
| Employee attrition | Engagement survey scores, manager 1:1 completion rate |
| Delivery success rate | Order fulfilment time, stock availability rate |
Interview signal: analysts who only track lagging indicators are telling you what happened — not helping you prevent problems. A good analyst pairs lagging KPIs with leading indicators.
E-commerce and Retail KPIs¶
[Mid-level] Define Customer Acquisition Cost (CAC) and Customer Lifetime Value (CLV). Why does the ratio matter?
Show answer
CAC: Total marketing + sales spend / Number of new customers acquired in the same period.
CLV: Predicted total revenue a customer will generate over their relationship with the business.
CLV (simple) = Average Order Value × Purchase Frequency × Customer Lifespan
CLV (margin) = CLV × Gross Margin
The ratio (CLV:CAC): - < 1: You're spending more to acquire a customer than they'll ever generate. Not viable. - 1-3: Marginal. You're barely covering acquisition costs. - 3-5: Healthy. Industry benchmark for most e-commerce businesses. - > 5: Potentially under-investing in growth. You could spend more on acquisition and still be profitable.
Why it matters for interviews: CLV:CAC is the lens for evaluating marketing efficiency. A company spending £200 to acquire a customer worth £600 is in a very different position from one spending £200 for a £250 customer.
[Mid-level] What is churn rate? How do you calculate it and what are its limitations?
Show answer
Churn rate = customers lost in a period / customers at the start of the period × 100
churn_rate = (customers_start - customers_end + customers_acquired) / customers_start
# Simpler version:
churn_rate = customers_lost / customers_at_start
Example: 500 customers at start of month, 40 churned, 60 new acquired = 480 at end. Churn rate = 40/500 = 8% monthly.
Limitations: 1. Doesn't distinguish segments: 8% churn among VIP customers is catastrophic; 8% among trial users is expected. 2. Revenue churn matters more than customer churn: losing 5 small customers (5% customer churn) but retaining 2 enterprise customers means revenue actually grew. 3. Doesn't capture at-risk customers: someone who hasn't churned yet but is disengaged isn't counted.
Better metrics: Monthly Recurring Revenue (MRR) churn, Net Revenue Retention (NRR), product engagement signals.
[Mid-level] What is conversion rate and how do you improve it?
Show answer
Conversion rate = (Number of conversions / Total visitors) × 100
The denominator and numerator depend on what "conversion" means: - E-commerce: purchase / sessions - SaaS: trial sign-up / landing page visits - Email: click / send
How to investigate: look at the funnel. Where are users dropping off?
The biggest drop (68% → 42%) is the priority. Is it product detail page quality, price, or trust signals?
How to improve: A/B test specific changes at each funnel step — don't change everything at once (can't attribute improvement).
[Senior] Your CEO asks for "one metric to run the business." You only get one. What do you choose for an e-commerce business and why?
Show answer
There's no single right answer — but the interviewer wants to see your thinking. A strong answer for e-commerce:
Net Revenue Retention (NRR) (also called Net Dollar Retention)
- NRR > 100%: Existing customers are spending more — expansion revenue exceeds churn
- NRR < 100%: Churn is eating into existing customer revenue
Why this metric: - Combines churn, upsell, and downsell into one number - If NRR > 100%, the business grows even without acquiring any new customers - Forces focus on both retaining customers AND growing their value - More revealing than raw revenue (which can grow purely from acquisition while existing customers churn)
What I'd monitor alongside it: CAC payback period (to ensure growth is sustainable) and gross margin (to ensure revenue growth is profitable).
Operational KPIs¶
[Mid-level] What KPIs would you use to measure the performance of a customer support team?
Show answer
| KPI | Formula | Why it matters |
|---|---|---|
| First Contact Resolution (FCR) | % tickets resolved on first contact | Drives customer satisfaction + reduces cost |
| Average Handle Time (AHT) | Total handle time / total contacts | Efficiency measure — but don't optimise in isolation |
| Customer Satisfaction (CSAT) | Average satisfaction score post-contact | Direct feedback |
| Net Promoter Score (NPS) | % Promoters - % Detractors | Predicts referral behaviour and loyalty |
| Ticket Backlog | Open tickets > SLA threshold | Operational health |
| SLA Compliance | % tickets resolved within SLA | Contractual requirement |
Interview signal: mention that FCR and CSAT must be measured together. High AHT might be acceptable if FCR is also high (agent solves complex issues on first contact). Low AHT with low FCR means agents are rushing through tickets and creating repeat contacts.