Why post-paid billing logic fails for prepaid voice
Billing records what happened. Prepaid enforcement has to prevent what should not happen. A post-paid system assumes you can invoice later. A prepaid product cannot, which is why the cut-off decision belongs inside the live call.
A lot of voice platforms were designed around a single assumption: usage is measured, then it is invoiced. That model works when there is a contract behind the customer and a billing run at the end of the month. The system can afford to be patient, because the money is collected after the fact.
Prepaid voice removes that patience. The customer has already paid, the balance is finite, and once it is gone there is no invoice to fall back on. Reusing post-paid billing logic for a prepaid product quietly inherits an assumption that no longer holds, and the gap shows up as lost credit.
What post-paid assumes
Post-paid logic is built to record and reconcile. Each of these assumptions is reasonable when there is an invoice at the end of the cycle to carry the risk.
You can trust the customer and reconcile usage later, because the relationship and the contract carry the risk.
Soft caps and alerts are enough, because a warning email or a flagged account is an acceptable first response.
A negative balance is recoverable, because anything consumed beyond expectation can be added to the next invoice.
Rating can run at the end of the cycle, because the only deadline that matters is the billing run, not the call.
Why that breaks on prepaid
Take away the invoice and every one of those assumptions turns into a loss the operator absorbs directly.
Uncollectable negative balances
There is no invoice to chase. If a prepaid balance is allowed to go below zero, the operator has simply given away minutes it will never be paid for.
Disputes over consumed minutes
When rating happens after the call, the customer can burn through credit they did not have. Arguing about minutes already spoken is a support cost with no clean resolution.
Fraud on stolen or shared PINs
A leaked or shared PIN with no live limit is an open line. By the time after-the-fact rating notices the spike, the credit is already gone and so is the caller.
Reconciliation overhead and write-offs
Every gap between what was consumed and what was funded becomes manual reconciliation, a write-off, or both. The cost lands on margin, not on the customer.
What prepaid actually needs
Prepaid voice needs the spending decision to live inside the call, not in a billing run that arrives long after the money is gone.
Authorise at call setup
Before the call connects, the balance is checked and a decision is made. A call that cannot be paid for should never be allowed to start.
Decrement the balance live
Per-second rating against the A to Z rate deck runs while the call is in progress, so the remaining balance always reflects the conversation as it happens.
Hard cut-off at zero
The moment credit reaches zero, the call ends. Not a soft cap, not an alert, not a flag for later review. The enforcement is the disconnect itself.
Awareness of concurrent calls
One balance can carry several simultaneous calls. Enforcement has to account for every active session against that credit at once, or the limit means nothing.
Two different jobs, not one delayed job
None of this means billing is wrong. Rating, CDRs, and reconciliation are still essential for reporting, settlement, and audit. The point is that they answer a different question. Billing tells you what a call cost. Enforcement decides whether a call may continue.
Seshnova is built around the enforcement side: authorise at setup, rate per second against the A to Z rate deck, decrement the balance live, account for concurrent calls on one balance, apply fraud caps, and hard cut-off the moment credit reaches zero. The CDRs and rating still produce the record afterwards, because both jobs matter. They are just not the same job.
Billing records calls. Enforcement stops them. A prepaid product needs both, and it cannot fake the second with the first.