Modeling lifetime value for postpaid, prepaid and IoT lines

A marketing team runs one LTVLTVLifetime Value: the total revenue (or profit) a customer generates throughout their entire relationship with your business.View full definition → number across the whole base: $1,850. That single figure sets the acquisition budget for a family postpaid plan, a $20 prepaid top-up user and a connected car SIM. Those three customers share no margin structure, no tenure, and no common definition of leaving. Blend them and you overspend on the cheapest segment while starving the most valuable one. So: three models, and a clear view of where each one breaks.

Why one blended LTV breaks

LTV is the gross profit you expect from a customer across the whole relationship, discounted for time and weighted by the probability they are still connected.

The generic formula:

LTV = (ARPU × Gross MarginGross MarginGross margin is the share of revenue left after subtracting the direct cost of producing goods or services, expressed as a percentage of revenue.View full definition → %) ÷ Churn RateChurn RateChurn rate is the percentage of customers or revenue lost over a period. It measures how fast a business loses its existing customer base.View full definition →

ARPU is monthly revenue per line, margin strips out network and service cost, and churn is the monthly departure probability you already measure the way the retention lesson sets out. Where each input sits against published comparators belongs to the benchmarking lesson (carrier 10-Ks and FCC reports among the sources it works through). Every number below is an illustrative input, chosen to show the shape of a model rather than to stand as a market figure.

One formula, one hazard rate, one revenue line. That works on a homogeneous base. Postpaid, prepaid and IoT differ on every input, and on the arithmetic's buried assumption: that the monthly probability of leaving is the same in month 2 and month 62. It almost never is.

Postpaid: high ARPU, sticky, contract-bound

Postpaid customers pay monthly, often under device financing, frequently on a multi-line account.

Worked example:

• ARPU: $50/month
• Gross margin: 60%
• Monthly churn: 1.2% (implied average tenure ≈ 1 ÷ 0.012 ≈ 83 months)

LTV = ($50 × 0.60) ÷ 0.012 = $30 ÷ 0.012 = $2,500

Now the failure mode. Dividing by a blended rate assumes a memoryless hazard. Real postpaid hazard is bimodal: elevated in the first 90 days (coverage disappointment, mis-sold plan, port-back to the previous carrier) and again the month the device instalment plan ends. Consider two cohorts. Cohort A loses 12% in year one, then settles at 1.2% a month. Cohort B is locked flat for 24 months, then 30% walk at payoff, and the survivors run at 1.2%. Both land within a few dollars of $2,500 undiscounted. They are not the same asset: cohort B has a subsidised handset outstanding for two years and reaches CACCACCustomer Acquisition Cost (CAC) is the total sales and marketing spend divided by the number of new customers gained in a period. It measures how efficiently you grow.View full definition → payback roughly a year later than cohort A. The blended rate cannot see that, and payback month is what a CFO gates spend on. Build LTV as the sum of monthly margin × survival probability from a cohort curve, not as one division.

Family plans need their own treatment. A fourth line on an existing account carries near-zero incremental acquisition cost and lower standalone ARPU, but it raises household stickiness, because switching means moving everyone. Model added lines as an increment to household LTV, never as standalone customers.

Prepaid: low ARPU, no contract, volatile tenure

Prepaid users pay in advance, with no credit check and no commitment. Nobody cancels; they just stop recharging, so operators declare a line churned after a dormancy window, commonly 60 to 90 days without a top-up.

Worked example:

• ARPU: $32/month
• Gross margin: 50%
• Monthly churn: 4% (implied tenure ≈ 25 months)

LTV = ($32 × 0.50) ÷ 0.04 = $16 ÷ 0.04 = $400

Roughly one-sixth of the postpaid figure, which is why a CAC ceiling borrowed from a blended LTV quietly destroys value here.

Two prepaid-specific traps. The first is recycling: a dormant user who comes back four months later registers as a fresh gross add on a new SIM, which inflates your acquisition count, truncates measured tenure and double-counts the same human. Stitching top-up events, app logins and device identifiers back to one profile is the only fix, and it is tooling work (Segment, which sells customer data platformcustomer data platformSoftware that unifies customer data from every source into one persistent profile that marketing, sales and service teams can act on.View full definition → software, exists largely to do this kind of identity resolution). Second: cohorts with a single recharge can carry negative LTV outright, once a bonus first month or a discounted handset is loaded in. Floor those cohorts at zero and price acquisition off the second-recharge cohort only.

The geography trap is larger still. Reliance Jio reports prepaid ARPU in the ₹180 to ₹200 range, a couple of dollars a month. Hold churn at 4% and margin at 50% and per-line LTV lands near $25. No paid search bid, no outboundoutboundProactive outreach that pushes your message to targeted audiences through advertising, email, or direct prospecting, initiated by the seller rather than the buyer.View full definition → save call, no handset contribution survives that arithmetic, which is why Jio's answer is self-serve digital onboarding and its own distribution rather than commissioned acquisition. Add multi-SIM ownership, common across India and much of Africa, and one person holds two prepaid lines while treating the second as a data-only spare: per-line LTV halves, household revenue does not.

IoT and connected-device SIMs: low ARPU, extreme tenure, different margin logic

IoT lines cover connected cars, smart meters, asset trackers, POS terminals and industrial sensors. Vodafone alone reports well over 100 million IoT connections, and Verizon Business runs one of the larger enterprise IoT books in North America, on per-unit revenue that would look like a rounding error in consumer.

Worked example:

• ARPU: $4/month
• Gross margin: 70% (data volumes are tiny)
• Effective monthly churn: 0.3%, driven by contract events rather than individual disconnects

LTV = ($4 × 0.70) ÷ 0.003 = $2.80 ÷ 0.003 = ≈ $933

Close to the prepaid figure on one-eighth the ARPU. Tenure does all the work.

Except that 0.3% is a fiction of convenience. IoT churn is a step function, not a decay curve: a smart-meter fleet is flat for seven years and then goes to zero inside a quarter, when the enterprise retenders or the underlying radio disappears. Verizon retired its 3G CDMA network at the end of 2022, and every module without an LTE or Cat-M path stopped working on a date known years in advance. Model an IoT fleet as a fixed-term annuity to the module's end of life (72 months × margin for a six-year deal), then multiply the next term by an explicit renewal probability. A 2G or 3G sunset date inside your modelled horizon is a hard truncation, not a churn assumption.

Two more things ARPU hides. Enterprise contracts frequently carry volume ratchets, so per-SIM price falls as the fleet grows: success pushes ARPU down. And a real share of shipped SIMs sit installed but never commissioned, earning nothing while still consuming platform cost. Acquisition also happens per contract, so a single deal can add hundreds of thousands of lines, which makes per-connection CAC meaningless and per-fleet CAC the only number worth reporting.

Building a three-line model

One spreadsheet, three tabs, one shared structure.

Segment      | ARPU  | Margin% | Monthly Churn | LTV
Postpaid     | $50   | 60%     | 1.2%          | $2,500
Prepaid      | $32   | 50%     | 4.0%          | $400
IoT (per SIM)| $4    | 70%     | 0.3%          | $933

Then set segment-specific CAC ceilings, using the loaded acquisition figures the CAC lesson builds rather than media cost. Holding LTV:CAC above 3:1 is a heuristic imported from SaaS marketing and worth treating as directional only:

• Postpaid: ceiling ≈ $833
• Prepaid: ceiling ≈ $133
• IoT: ceiling ≈ $311 per SIM, but assessed per enterprise contract across thousands of lines

That gap is why prepaid acquisition lives on retail top-up displaysdisplaysThe total number of times an ad or piece of content is displayed, regardless of clicks. Each display counts as one impression, even to the same person.View full definition → and SMS, while postpaid can carry device subsidies and staffed retail.

Adjusting for discounting and cross-sell

1. Discount future cash flows. A customer worth $2,500 over 83 months is not worth $2,500 today. Telecom finance teams typically model with an annual rate in the high single digits to low teens, specific to their own cost of capitalcost of capitalThe blended rate a company pays to finance itself through debt and equity. It sets the minimum return an investment must clear to create value.View full definition →. Long-tenure segments take the hardest hit: a 72-month IoT annuity discounted at 10% a year is worth about 55 months of margin, roughly three-quarters of face value, and the back-loaded postpaid cohort from the earlier example loses more than the front-loaded one despite the identical undiscounted total.

2. Layer in cross-sell and upsell. Postpaid LTV should carry the probability of an unlimited-data upgrade, a content bundle or an added family line. Vodafone and other converged European operators model household LTV across mobile, broadband and TV rather than single-product LTV, since converged customers churn at materially lower rates in their own reporting.

🎬 [VIDEO: "Customer Lifetime Value Explained" - https://www.youtube.com/results?search_query=customer+lifetime+value+explained+telecom - a primer on CLV mechanics applicable across subscription industries, useful for reinforcing the core formula before applying it to telecom segments]

Key takeaways

• Never run one blended LTV across postpaid, prepaid and IoT: the inputs differ by multiples, and a shared number misallocates acquisition budget in both directions.
• Postpaid LTV comes from high ARPU and long tenure, prepaid LTV is capped by dormancy churn, and IoT LTV is almost entirely a tenure story on tiny per-line revenue.
• Dividing margin by a blended churn rate assumes constant hazard. Postpaid hazard spikes at onboarding and at device payoff, so two cohorts with the same LTV can differ by a year on payback.
• Model IoT as a fixed-term annuity to module end of life with an explicit renewal probability, and treat any 2G or 3G sunset inside the horizon as a hard truncation.
• Prepaid needs identity stitching so returning users are not recounted as gross adds, a zero floor on single-recharge cohorts, and a rebuilt model market by market: at Jio-level ARPU, per-line LTV lands near $25 and most acquisition tactics stop clearing the bar.