The telecom value tree
Telecom economics are a base of subscribers, a spread per subscriber and a very large fixed asset. Small movements in churn or ARPU move more value than almost anything on the cost line.
01 The same tree, this industry
Where the money is made and lost here
The structure does not change: value is profit plus how well that profit becomes cash, profit is revenue minus cost, and revenue is price times quantity. What changes is which drivers sit underneath each branch, and which system holds them. If you have not read the general version, start with the Enterprise Value Tree and come back.
This tree is dominated by a denominator. Nearly every driver is expressed per subscriber, so the subscriber count is load-bearing for the whole model — and it is usually different in billing, CRM and the network inventory.
Every driver below names the metric it lands on. Follow one and you get its formula, the system the number lives in, and the ways it is commonly misread.
02 Price
What decides the price you actually get
ARPU by plan
The headline, and an average that hides the two segments actually moving: new cohorts and out-of-contract legacy.
Average net price net_revenue ÷ unitsPriceRetention offers and discounting
The price paid to keep a subscriber, granted at the point of churn risk and rarely measured against the value retained.
Price realisation net_revenue ÷ list_revenuePriceDevice subsidy
A commercial cost recognised as a price concession, spread across a contract term and often financed.
Contribution margin (net_revenue − variable_cost) ÷ net_revenuePriceAdd-ons and attachment
Insurance, content, extra data and roaming packs — the margin-rich part of the bill.
Attach rate orders_containing_the_attached_item ÷ eligible_ordersPriceInterconnect and roaming
Wholesale rates that move with agreements rather than with customers, and settle late.
Average net price net_revenue ÷ unitsPriceB2B contract pricing
Bespoke, multi-service and rarely reconciled against the service actually delivered.
Price realisation net_revenue ÷ list_revenue03 Quantity
What decides how much you sell
Subscribers and net adds
Gross adds minus churn, by segment. The base is the denominator under every other number in this tree.
Active customers count(distinct customers with at least one purchase in the window)VolumeChurn
The single most valuable number in the sector, and the one most sensitive to how a subscriber is defined.
Customer retention rate customers_retained ÷ customers_at_startVolumeContract renewal and upgrade
The moment value is either extended or discounted away, decided in a retention conversation.
Net revenue retention (starting_revenue + expansion − contraction − churn) ÷ starting_revenueVolumeCoverage and addressable base
What the network can actually sell to, which is a capital decision made years earlier.
Capacity utilisation actual_output ÷ practical_capacityVolumeB2B win rate
Longer cycles, larger contracts, and a pipeline that behaves nothing like consumer.
Win rate opportunities_won ÷ opportunities_closedVolumeUsage growth
Data consumption that drives both revenue and the capex that serves it.
Purchase frequency orders_in_period ÷ active_customers04 Cost
What it takes to operate
Network operating cost
Sites, power, transmission and leases — largely fixed, and the reason scale matters so much here.
Energy per unit energy_consumed_kWh ÷ units_producedCostCapex and spectrum
The defining cost of the sector, committed years ahead of the revenue it enables.
Capex intensity capital_expenditure ÷ net_revenueCostField maintenance
Truck rolls and repair time, where the difference between planned and unplanned work is most of the cost.
Mean time to repair total_downtime_hours ÷ number_of_failuresCostCustomer service cost
Calls per subscriber per month, which is a product and network quality metric wearing a service costume.
Cost to serve (delivery + service + returns + order_handling) ÷ customers or ordersCostAcquisition and dealer commission
Paid up front, recovered over a contract term that churn can cut short.
Customer acquisition cost (sales_cost + marketing_cost) ÷ new_customers_wonCostBad debt
Concentrated in specific acquisition channels and plan types, and visible there long before it reaches the ledger.
Days sales outstanding (accounts_receivable ÷ net_revenue) × days_in_period05 Cash
Where the cash actually sits
Device financing receivables
A telco balance sheet increasingly looks like a lender's, and the receivable behaves like one.
Days sales outstanding (accounts_receivable ÷ net_revenue) × days_in_periodCashHandset inventory
Fast-depreciating stock with a launch calendar that no forecast fully survives.
Days inventory outstanding (inventory_value ÷ cogs) × days_in_periodCashCapex phasing
Cash out for coverage and capacity long before the subscribers arrive.
Capex intensity capital_expenditure ÷ net_revenueCashLeases and tower obligations
Long-dated commitments that sit between cost and capital structure.
Net debt to EBITDA (total_debt − cash_and_equivalents) ÷ EBITDACashLeverage
The sector runs on debt, so the covenant headroom is a strategic constraint rather than a finance detail.
Net debt to EBITDA (total_debt − cash_and_equivalents) ÷ EBITDACashReturn on capital employed
The only measure that judges a network investment against what it consumed.
Return on capital employed EBIT ÷ (total_assets − current_liabilities)06 Where the numbers live
The systems behind the branches
A value tree is only usable once each box maps to a system and a record. These are the six that matter most in telecom: what each one is actually for, the records inside it the tree depends on, and which branch it feeds.
| System | What it holds | Key records | Feeds |
|---|---|---|---|
| BSS / billing | Rating, charging and the invoice: what the customer was actually billed, by product and period. | Subscription, Rate plan, Charge, Invoice, Payment | Price, quantity, cash |
| CRM | The relationship, the interactions and the retention offers granted during them. | Customer, Account, Interaction, Offer, Case | Price, quantity |
| OSS / network inventory | What is deployed where, and which service runs over which element — the link between a fault and a customer. | Site, Element, Circuit, Service instance | Cost, quantity |
| Mediation and charging | Raw usage records before they become revenue: the highest-volume data in the business. | CDR, Session, Usage record | Price, quantity |
| Field service | Work orders, truck rolls and repair times, joined to the asset that failed. | Work order, Technician, Asset, Failure code | Cost |
| Network performance | Alarms, degradation and quality by cell and region — the leading indicator of churn nobody joins to churn. | Alarm, KPI sample, Cell, Region | Cost, quantity |
Almost every hard question in telecom needs two of these joined. That join — not the calculation — is the work.
07 Where it leaks
Value lost between two systems
These are the losses that no single system can see, because the evidence is split across two of them. Each one is a real number that stays invisible until the join exists — which is why the tree is an integration exercise before it is an analysis one.
| Where value leaks | Why it happens | The join that finds it |
|---|---|---|
| The subscriber who is three subscribers | A line in billing, a contact in CRM and a service in network inventory. ARPU, churn and cost per subscriber all divide by a number that depends which system you asked. | Resolve subscription, account and service instance to one subscriber record |
| Network quality never joined to churn | The cells with the worst experience and the postcodes with the highest churn are both known, in different systems, by different teams. | Join network KPI samples by cell to churn events by customer location |
| Commission paid on subscribers who left | Acquisition is paid at activation; churn happens months later. Channel profitability is measured on gross adds because the join is never made. | Join dealer commission to the subscriber lifetime that followed it |
| Retention offers with no value test | Discounts are granted at the moment of risk without knowing the subscriber's contribution, so the most valuable and least valuable customers are saved on identical terms. | Join the retention offer to lifetime contribution and remaining term |
| B2B service delivered, not the one billed | Enterprise contracts drift: circuits are added, moved and ceased faster than the billing record follows. | Reconcile the service inventory against the contracted and billed service, continuously |
08 Worked example
Net adds up, value down
The base grew and service revenue fell. Splitting the tree shows why: growth came from a low-ARPU channel, retention discounting repriced part of the existing base, and churn concentrated in cells with a known quality problem that no commercial report had ever seen.
| Component | Effect | What sits behind it |
|---|---|---|
| Gross adds | +€4.6m | Weighted to a lower-ARPU acquisition channel |
| Churn | −€5.2m | Concentrated in three regions with known network issues |
| Retention discounting | −€2.4m | Offers granted without a value test |
| Add-on attachment | +€1.1m | The one branch that improved |
| Bad debt | −€0.7m | Concentrated in one acquisition channel |
| Net | −€2.6m | A larger base earning less than the smaller one |
Illustrative figures, shown to demonstrate the split. The point is the shape of the walk, not the numbers — on your own data the same bridge is built from your ledger.
09 Diagnostics
Six questions to ask in telecom
Ask them of your own team before anyone asks them of you. In most organisations at least two of these cannot be answered without a manual exercise, and those two are the plan.
- How many subscribers do you have — and do billing, CRM and network inventory agree?
- Can you see churn by network experience, not just by tariff and tenure?
- What is the contribution of the subscribers you discounted to retain last quarter?
- Which acquisition channels produce subscribers still profitable after 18 months?
- For enterprise customers, does the service you bill match the service you deliver?
- Can you value a truck roll in contribution rather than in cost?
10 The metrics behind it
Definitions for every box
11 Questions
Frequently asked
Why is the subscriber count so contentious?
Because a subscription, an account, a SIM and a service instance are four different objects and each system counts the one it owns. ARPU, churn and cost per subscriber all divide by it, so a definition disagreement propagates into every number on the page.
What is the highest-value join in telecom?
Network experience to churn. Both datasets exist, both are large, and almost nobody joins them — which means the commercial team manages churn with tariffs while the cause sits in the radio network.
How should retention offers be governed?
By contribution and remaining term, decided at the moment of the offer. Any other basis saves the least valuable subscribers on the same terms as the most valuable ones.
Does this tree fit a tower or infrastructure business?
Partly. Revenue becomes tenancy rather than subscription, so the quantity branch changes entirely, but the capital and maintenance branches are, if anything, more dominant.
See this tree on your own data
Connect the systems above, define each box once, and the tree stops being a slide.