A–Z rate decks without the spreadsheet chaos
Thousands of destination prefixes, overlapping zones, peak and off-peak windows, connection fees, and per-route increments. An A–Z deck has a lot to get right, and a spreadsheet quietly gets it wrong.
An A–Z rate deck starts clean. Then it lives for years. New destinations are added, prices change, a reseller needs a special rate, and a quick fix is pasted in during a busy week. Over time the deck drifts: overlapping prefixes, stale destinations, duplicate labels, and manual overrides that nobody fully understands.
None of that is visible until a call is rated wrong. A spreadsheet will happily store an ambiguous prefix or a blank rate, because a spreadsheet does not know what a rate deck is supposed to mean. Managing A–Z rating well is mostly about removing that silence.
What a rate deck has to get right
A deck is not just a list of prices. It is the set of rules that turns a dialled number into a charge, and every one of those rules has to be precise.
Longest-prefix matching
A rate deck has to resolve every dialled number to exactly one rate. That means deterministic longest-prefix matching and a clear policy for overlapping prefixes, so a more specific destination always wins over a broader one.
Zones and destination labels
Prefixes roll up into zones, and zones carry the destination labels operators and resellers actually read. Labelling has to stay consistent, or the same destination ends up named three different ways across the deck.
Increments and minimum charges
Different routes bill differently. Per-route billing increments, minimum charges, and the rounding rules behind them all need to be explicit, not assumed, so per-second rating produces the figure each route expects.
Time windows and connection fees
Peak and off-peak windows, day-of-week variations, and per-call connection fees change the effective price of a route. These rules belong in the deck itself rather than in a separate note nobody updates.
Effective dates and versioning
Rates change. A deck needs effective dates and version history so a price increase applies from the agreed moment, and so anyone can see what a route cost on the day a call was rated.
Testing before go-live
Before a deck goes live it should be tested against sample calls, so unexpected matches, missing zones, and surprising charges surface in rehearsal rather than on a customer invoice.
Why spreadsheets fail at scale
A spreadsheet is fine for a handful of destinations. At A–Z scale, the things it cannot do become the things that cost money.
No validation, so a malformed prefix, a blank rate, or a typo in an increment loads without a single warning.
Silent overlaps, where two prefixes both match a number and nobody can predict which one rating will pick.
No version history, so a rate that changed last month cannot be reproduced when a dispute is raised today.
Copy-paste errors that shift a column, duplicate a row, or quietly drop a destination from the deck.
No safe way to test, so the only place a new deck is exercised is live production traffic.
What managed rating gives you
Treating the deck as managed data rather than a file turns the same work into something repeatable, checkable, and safe to change.
Deterministic matching
Every number resolves through the same longest-prefix logic, so overlapping prefixes follow a defined rule rather than the order rows happen to sit in a sheet.
Validation before publish
A deck is checked before it goes live. Missing zones, broken prefixes, duplicate labels, and absent rates are flagged while they are still cheap to fix.
Versioned, dated changes
Rate changes carry effective dates and a version record, so each price applies from the right moment and the deck that rated any historical call can be recovered.
Dry-run against sample calls
A new deck can be run against sample calls before publication, so the charges, increments, and connection fees are confirmed on representative traffic first.
Rate decks that stay under control
Seshnova rates A–Z traffic in real time, per second, against managed decks rather than loose spreadsheets. Matching is deterministic, changes are versioned with effective dates, and a deck can be dry-run against sample calls before it ever touches live traffic. The same rating feeds CDRs, balance and PIN control, and hard cut-off enforcement, so the figure on the deck is the figure that protects the balance.
If you run A–Z rating across multiple brands or resellers, the goal is the same: one set of rules, validated before publish, with a clear history of what changed and when.