Data Exercise: Analyze Goalhanger’s Subscriber Growth and Revenue — Practice Problems

Hook: Turn real media numbers into classroom-ready statistics practice

Students and teachers: tired of abstract textbook problems that never connect to real-world data? Use Goalhanger’s public subscriber and revenue figures to teach core analytics—growth rates, ARPU, churn and forecasting—so learners practise with the same metrics media managers use in 2026. These problems are ready for a 50–90 minute class, a homework worksheet, or an online mock exam.

Why Goalhanger (and why 2026)?

In January 2026 Press Gazette reported Goalhanger exceeds 250,000 paying subscribers and generates about £15m yearly subscriber income (average £60 per subscriber per year). That simple headline contains multiple teachable statistics: verifying ARPU, verifying revenue, modelling growth, and exploring churn and lifetime value. These practice exercises also let you introduce modern 2026 trends—subscription-first monetisation in podcasting, AI-driven personalisation to reduce churn, and scenario planning under post-2025 economic pressures.

Source: Press Gazette, Jan 2026 — "Goalhanger exceeds 250,000 paying subscribers".

Key formulas & concepts (classroom cheat sheet)

• ARPU (Annual) = Total annual subscription revenue / Number of subscribers
• Monthly ARPU = ARPU / 12
• Churn rate (period) ≈ Number of cancellations during period / Subscribers at start of period
• Average lifetime (periods) ≈ 1 / churn rate
• LTV (Revenue-based) = ARPU per period × Average lifetime (in same periods)
• CAGR = (Ending value / Starting value)^(1/years) − 1
• Simple forecast = last period value × (1 + growth rate)^n

How to use this worksheet in class

1. Start with the headline data (250k subs, £15m revenue). Ask students to verify ARPU and check rounding.
2. Split the class into groups. Assign one growth, one churn/LTV, one forecasting, and one hypothesis-testing problem.
3. Allow students to choose assumptions for monthly vs annual pricing where needed—encourage showing how results change with assumptions.
4. Finish with a short reflection: how do 2026 trends (AI personalisation, subscription fatigue, macro conditions) change your model?

Practice Problems — Classroom-ready (with data and clear tasks)

Problem 1 — Verify ARPU and revenue

Press Gazette reports: 250,000 paying subscribers and ~£15m annual subscriber income. Verify the arithmetic and compute monthly ARPU.

Task:

1. Compute ARPU = annual revenue / subscribers.
2. Compute monthly ARPU.
3. If 50% of subscribers pay monthly and the other 50% pay annually, what is the implied average monthly price for monthly payers if annual payers pay exactly £60 per year?

Problem 2 — Growth rates and CAGR

Assume Goalhanger had 120,000 paying subscribers on 1 Jan 2025 and 250,000 on 1 Jan 2026.

Task:

1. Compute the absolute increase and percentage growth over the year.
2. Compute the CAGR for that one-year period (confirm it equals the year-over-year growth).
3. If growth continues at the same annual rate for three years, what is the subscribers forecast on 1 Jan 2029?

Problem 3 — Monthly churn from cohort retention

Suppose a cohort of 10,000 subscribers is acquired in January. After 6 months, 6,000 remain. Assume constant monthly churn.

Task:

1. Estimate the monthly churn rate.
2. Estimate average subscriber lifetime in months.
3. Using ARPU = £60 per year, compute LTV in £ (use lifetime in years).

Problem 4 — Three forecasting scenarios (conservative, base, aggressive)

Using the reported 250,000 subs on 1 Jan 2026, build a 3-year revenue forecast under three scenarios:

• Conservative: flat subscribers (0% growth), ARPU falls 1% annually due to discounting and competition.
• Base: subscribers grow 15% year-over-year, ARPU stays at £60.
• Aggressive: subscribers grow 30% year-over-year for year 1 and 20% thereafter, ARPU improves 5% per year due to upsells and tiered pricing.

Task: Produce year-by-year subscribers and revenue for 2026–2028 under each scenario.

Problem 5 — Sensitivity and break-even churn

Goal: management wants at least £18m revenue in 2027. Starting from 250,000 subscribers and current ARPU £60, assume the company can acquire 40,000 new subscribers each year but also experiences churn. Compute the maximum annual churn rate (applied continuously each month, approximated) that keeps revenue ≥ £18m in 2027 under constant ARPU.

Problem 6 — Hypothesis test on ARPU change (2025 vs 2026)

Class data: sample of 200 subscribers in 2025 had mean annual payment £58 (sd £10). A sample of 200 subscribers in 2026 had mean £62 (sd £12). Test at α=0.05 whether ARPU increased.

Problem 7 — Quick regression: revenue drivers

Using monthly data for 12 months with two columns—subscribers (S_t) and ARPU monthly (A_t)—fit a linear regression revenue_t = β0 + β1×S_t + β2×A_t + ε_t. Explain what β1 and β2 capture and how multicollinearity might appear if ARPU is strongly correlated with subscribers.

Answer Key & Step-by-step Solutions

Problem 1 — Solution

ARPU = 15,000,000 / 250,000 = £60. Monthly ARPU = 60 / 12 = £5 per month.

If 50% pay annually at £60 each (125,000 × £60 = £7,500,000), the remaining £7,500,000 must come from 125,000 monthly payers. Annual revenue per monthly payer = 7,500,000 / 125,000 = £60 per year, so implied monthly price = £60 / 12 = £5. This indicates the simple split in the Press Gazette note implies parity between monthly and annual revenue per subscriber (many publishers price annual at a discount, but here the headline averages to £60).

Problem 2 — Solution

Absolute increase = 250,000 − 120,000 = 130,000 subscribers. Percentage growth = 130,000 / 120,000 = 1.0833 = 108.33% growth year-over-year.

CAGR over one year = (250,000 / 120,000)^(1/1) − 1 = 108.33%.

Three-year forecast at same annual rate: subscribers_2029 = 250,000 × (1 + 1.0833)^3. Numerically: (1 + 1.0833) = 2.0833; 2.0833^3 ≈ 9.03; subscribers ≈ 250,000 × 9.03 ≈ 2,257,500. Note: such large extrapolation shows why exponential forecasts must be grounded in market saturation assumptions.

Problem 3 — Solution

Retention after 6 months = 6,000 / 10,000 = 0.6. With constant monthly retention r = (1 − churn), r^6 = 0.6. Thus r = 0.6^(1/6) ≈ 0.912. Monthly churn = 1 − r ≈ 0.088 = 8.8% per month.

Average lifetime in months = 1 / churn ≈ 1 / 0.088 ≈ 11.36 months ≈ 0.95 years.

LTV = ARPU per year × average lifetime in years = £60 × 0.95 ≈ £57.

Problem 4 — Solution (year-by-year numbers)

Base numbers: year 0 (2026) subscribers = 250,000; ARPU = £60.

• Conservative: subs constant 250,000. ARPU: 2026 £60, 2027 £59.40 (−1%), 2028 £58.81. Revenue 2026 = 250,000×60 = £15,000,000; 2027 = 250,000×59.40 = £14,850,000; 2028 = 250,000×58.81 ≈ £14,702,500.
• Base (15% y/y sub growth): subscribers 2026=250k, 2027=287,500, 2028=330,625. ARPU constant £60. Revenue 2026=£15m, 2027=287,500×60=£17,250,000, 2028=330,625×60=£19,837,500.
• Aggressive: subs grow 30% in 2027 then 20% in 2028 → 2027 subs=250k×1.30=325,000; 2028 subs=325,000×1.20=390,000. ARPU rises 5% each year: 2026 £60, 2027 £63, 2028 £66.15. Revenue 2026=£15m; 2027=325,000×63=£20,475,000; 2028=390,000×66.15≈£25,798,500.

Problem 5 — Solution (approximate)

We start 2026 with 250,000 subs. Each year Goalhanger acquires 40,000 new subs. Let churn rate per year be c. Subscribers at end of year: S1 = (250,000 × (1 − c)) + 40,000. Do two iterations to reach end of 2027 (two years total) to compute revenue in 2027 = S2 × £60. We need S2 × 60 ≥ 18,000,000 → S2 ≥ 300,000 subs.

Compute S1: S1 = 250,000(1 − c) + 40,000. S2 = S1(1 − c) + 40,000 = ((250,000(1 − c) + 40,000)(1 − c) + 40,000).

We want S2 ≥ 300,000. Solve numerically: Let’s test c = 0.2 (20%): S1 = 250k×0.8+40k=200k+40k=240k. S2 = 240k×0.8+40k=192k+40k=232k < 300k (fail).

Try c = 0.0: S1=290k, S2= (290k×1) + 40k = 330k → success. We need to find max c s.t. S2 ≥ 300k. Solve algebraically:

S1 = 250,000(1 − c) + 40,000

S2 = S1(1 − c) + 40,000 = [250,000(1 − c) + 40,000](1 − c) + 40,000

Set S2 = 300,000 and solve for c numerically. Expand: S2 = 250,000(1 − c)^2 + 40,000(1 − c) + 40,000 = 300,000.

Rearrange: 250,000(1 − c)^2 + 40,000(1 − c) − 260,000 = 0. Let x = (1 − c). Then 250,000 x^2 + 40,000 x − 260,000 = 0. Divide by 10,000: 25 x^2 + 4 x − 26 = 0.

Solve quadratic: x = [−4 ± sqrt(4^2 − 4×25×(−26))] / (50) = [−4 ± sqrt(16 + 2600)] / 50 = [−4 ± sqrt(2616)] / 50. sqrt(2616) ≈ 51.14. Positive root: (−4 + 51.14)/50 = 47.14/50 = 0.9428. Then c = 1 − x = 0.0572 ≈ 5.72% annual churn maximum.

So the maximum annual churn ≈ 5.7% if 40,000 new subs are acquired each year and ARPU stays constant.

Problem 6 — Hypothesis test solution

Null H0: μ2026 = μ2025; Alternative H1: μ2026 > μ2025. Two-sample t-test (assuming unequal variances approx).

Given n1 = n2 = 200, mean1 = 58, sd1 = 10; mean2 = 62, sd2 = 12. Test statistic t = (62 − 58) / sqrt( (10^2 / 200) + (12^2 / 200) ) = 4 / sqrt( (100/200) + (144/200) ) = 4 / sqrt(0.5 + 0.72) = 4 / sqrt(1.22) = 4 / 1.1045 ≈ 3.62.

Degrees of freedom ~ use Welch’s approximation (large n so df≈ ~350). Critical one-sided t at α=0.05 ≈ 1.645. Since 3.62 > 1.645, reject H0; there is statistically significant evidence ARPU increased.

Problem 7 — Regression guidance

In revenue_t = β0 + β1 × S_t + β2 × A_t + ε_t, β1 measures revenue change per additional subscriber holding ARPU constant; β2 measures revenue change per unit change in monthly ARPU holding subscribers constant. If S_t and A_t are highly correlated (e.g., when higher subscriber months are also higher-priced months), standard errors inflate and coefficient estimates become unstable. Teach variance inflation factor (VIF) diagnostics and encourage centring or principal component regression if multicollinearity arises.

Advanced classroom extensions (2026 trends & predictions)

Use these exercises to discuss 2026-specific trends:

• AI-driven retention: in late 2025 publishers increasingly use ML recommendations and personalised drip content; include a scenario where retention improves by 2–4% for targeted cohorts.
• Revenue diversification: podcast networks add premium newsletters and live events (tickets), which change ARPU composition—ask students to model ARPU as base subscription plus upsell revenue.
• Macro headwinds: post-2025 inflation and cost-of-living squeezes raised churn for some publishers—run sensitivity analysis for an economic shock increasing churn by 3–6 percentage points for 12 months.
• Privacy & measurement: with third-party cookies deprecated and ID solutions evolving in 2025–26, conversion rate estimates are noisier. Teach students to include larger confidence intervals or use Bayesian priors for acquisition forecasting.

Tools, Excel formulas & Python snippets for teachers

Excel formulas (examples):

• ARPU: =TotalRevenue / Subscribers
• Monthly churn from cohort: =1 - (Retention)^(1/Months)
• CAGR: =(End/Start)^(1/Years)-1
• Quadratic solver (for Problem 5): use the quadratic formula with SQRT()

# Python (numpy) example: compute monthly churn from cohort
import numpy as np
retention = 6000 / 10000
months = 6
r = retention ** (1/months)
monthly_churn = 1 - r
print('Monthly churn:', monthly_churn)

Classroom tips to deepen learning

• Ask students to produce charts: subscriber curve, revenue scenarios and cohort retention curves.
• Make them defend assumptions: why pick 15% growth vs 30%? What market signals justify each?
• Have groups present sensitivity tables: show how a ±2% ARPU shock changes revenue vs a ±2% churn shock.
• Introduce ethics and trustworthiness: emphasise transparent assumptions and how mis-stated ARPU or hidden discounts can mislead stakeholders.

Quick classroom-ready worksheet (one-page)

1. Verify ARPU and monthly ARPU from headline figures. (2 marks)
2. Compute YoY growth and CAGR from 120k to 250k subs. (3 marks)
3. From a 10k cohort with 6k retention at 6 months, compute monthly churn and LTV. (4 marks)
4. Produce conservative/base/aggressive revenue numbers for 2026–2028. (6 marks)
5. Run a short hypothesis test comparing two ARPU samples. (5 marks)

Final thoughts — why hands-on media datasets matter in 2026

Students trained on realistic datasets leave class better prepared for roles in analytics, product, and media strategy. Goalhanger’s publicly reported numbers provide a tidy, credible starting point: the headline ARPU and subscriber totals let you teach a suite of industry-relevant techniques from cohort analysis to scenario-based forecasting. In 2026, with subscription-first strategies, AI-led retention, and evolving privacy rules, analysts must pair solid statistics with clear assumptions. These exercises practise both.

Call to action

Download the printable practice worksheet and Excel template tailored for these problems, or request the Python/Jupyter notebook with simulated cohorts and forecast models. Use the worksheet in your next class or mock exam—then share student results for feedback. Click to get the free teacher pack and join our educator community for more up-to-date media analytics problems (updated through 2026).

Related Reading

• Plug-and-Play Breakfast Soundtracks: Best Bluetooth Speakers Under $50 for Your Pancake Brunch
• Micro-App Marketplaces for NFT Utilities: How to Launch, List, and Price Small Apps
• When AI Chip Demand Raises Costs: How Rising Memory Prices Affect Travel Tech Budgets
• Metal Prices, Geopolitics and OTC Miners: Building a Commodity-Focused Penny Stock Scanner
• Will Shifting Asian Markets Change Tapestry Prices? What Sellers Need to Watch