# Number recycling and phone-list decay: how often to re-check your list

> Disconnected phone numbers are reassigned to new people, so every list decays. How recycling works, what the US rules say, and a re-check cadence recipe.

Canonical: https://mobilevalidate.com/blog/number-recycling-and-list-decay · Last updated: 2026-10-30

![A phone number moves around a recycling loop from one person to the next after a 45 to 90 day aging period, beside a bar chart of valid rows falling over time and a re-check timeline.](https://mobilevalidate.com/images/blog/number-recycling-and-list-decay.svg)

*Disconnected numbers are reassigned, so every list decays unless you re-check it.*


By MobileValidate team (https://mobilevalidate.com/about) · Published: 2026-10-30 · Category: Deliverability · Tags: Number recycling, Reassigned numbers, List hygiene, Bulk jobs, Deliverability

Phone numbers are recycled: when a customer disconnects, the operator later gives the number to someone new. Every stored list therefore decays, and the person behind a number you had consent for may no longer be the person you reach. Re-check lists on a schedule, compare each answer with your last snapshot, and re-confirm consent when a number changes.

## Why do phone numbers get recycled?

Numbers are a limited resource. Each operator holds blocks of numbers from its national numbering plan, and regulators expect those blocks to be used efficiently. When a subscriber cancels or stops paying, the number rests for a while and then goes back into the pool for new customers.

For a business, that creates three problems:

- **Wrong person.** A message meant for your customer reaches a stranger, which can breach consent rules and leak information.
- **Wasted spend.** Messages to disconnected numbers fail and still cost money or reputation.
- **Stale signals.** A number that once passed your checks may now belong to someone quite different.

None of this is visible in the number itself. `+1 202 555 01xx` looks the same before and after it changes hands.

## How long do numbers rest before reassignment?

It depends on the country, and on the type of customer.

In the United States, FCC rules define "aging numbers" as "disconnected numbers that are not available for assignment to another end user or customer for a specified period of time". Numbers of residential customers "may be aged for no less than 45 days and no more than 90 days". Numbers of business customers may be aged for 45 to 365 days ([eCFR, 47 CFR 52.15](https://www.ecfr.gov/current/title-47/chapter-I/subchapter-B/part-52/subpart-B/section-52.15)). In other words, a US consumer's disconnected number can be someone else's within a few months.

Other countries set their own rules, and the resting period for mobile numbers varies. Check the numbering regulator of each country you send to, or ask your messaging provider, before you assume a period.

## What does the US Reassigned Numbers Database do?

In 2018 the FCC decided to create a single database of permanently disconnected numbers, so callers could find out whether a number had been reassigned since they got consent. The order, published in 2019, also set a minimum aging period of 45 days and "a limited safe harbor from liability for any caller that relies upon inaccurate information provided by the database" ([FCC, 2019](https://www.federalregister.gov/documents/2019/03/26/2019-05620/advanced-methods-to-target-and-eliminate-unlawful-robocalls)). Subscriptions began on 1 November 2021 ([FCC, 2021](https://www.federalregister.gov/documents/2021/11/05/2021-23855/consumer-and-governmental-affairs-bureau-announces-interim-usage-charges-for-the-reassigned-numbers)).

Service providers report permanently disconnected numbers every month ([RND](https://www.reassigned.us/)). Callers query a number together with a date, typically when they obtained consent or last reached the person, and the database says whether the number "has been permanently disconnected since a date provided by the caller" ([FCC, 2019](https://www.federalregister.gov/documents/2019/03/26/2019-05620/advanced-methods-to-target-and-eliminate-unlawful-robocalls)). Our [Reassigned Numbers Database guide](/blog/reassigned-numbers-database-explained) explains the query, the safe harbor and how it fits next to other checks.

The database covers US numbers only. It answers one precise question (disconnected since date X?) and says nothing about line type, carrier or reachability.

## How fast do lists decay?

We won't give you a universal percentage. Published "your list loses N % a year" figures mix countries, list sources and definitions, so they rarely apply to your list. What you can measure is your own decay:

1. **Take a random sample** of the list, a few thousand numbers.
2. **Re-check it** with the same checks you used last time.
3. **Count changes**: numbers that became invalid, carrier or line type changes, messenger accounts that appeared or disappeared, and numbers that now fail delivery.
4. **Divide by the months** since the last check. That is your monthly change rate.

Different lists decay at very different speeds. A list of active customers who log in weekly barely moves. A list of trade-show leads from three years ago may have changed a lot. Measure each source separately.

The industry treats disconnection as an ongoing process, not a yearly clean-up. The US wireless industry's guidelines say senders "should process telephone deactivation files regularly (e.g., daily) and remove any deactivated telephone numbers from any opt-in lists" ([CTIA, 2023](https://api.ctia.org/wp-content/uploads/2023/05/230523-CTIA-Messaging-Principles-and-Best-Practices-FINAL.pdf)).

## Which signals suggest a number changed hands?

No single check says "reassigned" outside the US database. Changes between two snapshots are the evidence:

| Change since last snapshot | What it may mean | Suggested action |
|---|---|---|
| Number now fails validation | Numbering plan changed, or the data was wrong | Correct or remove |
| Carrier differs | Ported, possibly with a new owner | Treat as "verify before sending" |
| Line type changed (for example mobile → VoIP) | New service on the number | Re-confirm consent |
| Messenger account disappeared | Account deleted or number changed hands | Re-confirm before messaging |
| Messenger account appeared on a long-silent number | Possibly a new owner | Re-confirm before messaging |
| Repeated unknown-subscriber delivery errors | Disconnected | Remove, and check the RND for US numbers |

Messenger accounts don't disappear the moment a number is disconnected. Each app deletes inactive accounts on its own schedule, so an account can outlive the subscriber's contract. Treat presence as a helpful signal, not a record of ownership. Always keep `checked_at` with each answer: a fact is only as good as its date.

To put a price on that decay for one campaign, the free [SMS campaign savings calculator](/tools/sms-campaign-savings-calculator) compares the SMS spend on dead numbers with the cost of checking first.

## What is a good re-check cadence?

Match the cadence to how you use the list and how fast it decays. A starting recipe:

| List | When to re-check | Checks |
|---|---|---|
| Active customers (transactions in the last 90 days) | When a delivery fails, and yearly | Carrier |
| Marketing opt-ins in regular use | Monthly, and before large campaigns | Carrier plus the messenger you send on |
| Dormant opt-ins (no contact for 6+ months) | Before any send | Carrier, messenger, and the RND for US numbers |
| Purchased or old lead lists | Don't message until consent is confirmed | Validate and check before any outreach |
| Numbers used for account recovery | On login, if the last check is older than your window | Carrier (porting hints) |

Then tune it. If your measured change rate is low, stretch the interval. If it is high, shorten it or re-confirm consent more often.

## How do you run the re-check with bulk jobs?

A bulk job checks up to 50,000 numbers in the background, at the lower bulk price, with the original row order kept in the results. First, a free estimate:

```bash
curl https://api.mobilevalidate.com/v1/jobs/estimate \
  -H "Authorization: Bearer $MOBILEVALIDATE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"numbers": ["+447700900001", "+447700900002", "+447700900003", "+447700900002", "0123"],
       "checks": ["carrier", "whatsapp"], "max_age": 2592000}'
```

```json
{"object": "estimate", "total": 5, "valid": 3, "invalid": 1, "duplicate": 1, "cached": 0,
 "unsupported": 0, "suppressed": 0, "checks": ["network.carrier", "whatsapp.registered"],
 "checks_total": 10, "billable_max": 0, "max_cost": {"amount": "0", "currency": "USD"}}
```

(Test keys always estimate zero.) `max_age: 2592000` accepts cached answers up to 30 days old, which are free. For a strict re-check, use a shorter `max_age`, or `0` to force fresh checks, which are billed. Then create the job with the same body and an `Idempotency-Key`, and page through the results. In test mode, the three valid numbers come back like this:

```text
+447700900001  network.carrier: completed (mobile, Test Carrier)   whatsapp.registered: true
+447700900002  network.carrier: unknown (NO_DATA)                   whatsapp.registered: false
+447700900003  network.carrier: unknown (UPSTREAM_TIMEOUT)          whatsapp.registered: unknown (UPSTREAM_TIMEOUT)
```

`GET /v1/jobs/{id}/results?service=whatsapp&registered=false` returns only the rows without an account, and the CSV download keeps your original row order and adds the result columns. The [bulk jobs documentation](/docs/bulk-jobs) covers the full flow, including webhooks when the job completes.

## How do you compare with your last snapshot?

Store a small record per number after each run: the E.164 number, the carrier, the line type, the messenger results and `checked_at`. After the next run, compare field by field:

```js
// prev and next: {carrier, line_type, whatsapp, checked_at}; null means unknown.
function snapshotChanges(prev, next) {
  const changed = [];
  for (const k of ["carrier", "line_type", "whatsapp"]) {
    if (prev[k] == null || next[k] == null) continue;   // unknown on either side: no evidence
    if (prev[k] !== next[k]) changed.push(k);
  }
  return changed;                                        // e.g. ["carrier"] → re-confirm before sending
}
```

Skip comparisons where either side is unknown. Inconclusive answers (unknown, unsupported country, timeout) aren't charged, and they aren't evidence of change.

## What should you do with numbers that changed?

A change is a reason to stop and confirm, not a reason to keep sending. Handle each group deliberately:

- **Invalid or disconnected:** remove the number from sending lists. Keep a suppression record so a later import doesn't bring it back.
- **Possibly reassigned:** pause marketing messages. If the person still has another channel with you, such as an e-mail address or an app account, ask them to confirm their number there.
- **Changed but still plausible** (for example a port to another carrier): keep sending transactional messages the customer expects, and re-confirm at the next natural moment, such as a login.
- **No answer (unknown):** keep the number, keep the old snapshot, and try again at the next run.

Consent belongs to a person, not to a number. When the evidence suggests a new person holds the number, treat the old consent as not carrying over. Only check numbers you have a lawful reason to hold, and delete what you no longer need. Consent and reassignment rules differ by country, so treat this as general guidance and consult counsel for your markets.

## What are the key takeaways?

- Disconnected numbers are reassigned. In the US, residential numbers can rest for as little as 45 days.
- The US Reassigned Numbers Database answers "disconnected since my consent date?" for US numbers, with a limited safe harbor.
- Don't trust universal decay figures. Measure your own change rate by re-checking a sample.
- Re-check on a cadence that matches how each list is used, and before any send to dormant numbers.
- Compare each run with your last snapshot, keep `checked_at`, and re-confirm consent when a number changes.

For the mechanics of cleaning a whole file, read [how to clean a phone number list in bulk](/blog/how-to-clean-a-phone-number-list-in-bulk). For what failed deliveries tell you, see [why SMS messages aren't delivered](/blog/why-sms-is-not-delivered).

## Sources

1. [47 CFR 52.15: Central office code administration (number use categories, aging numbers)](https://www.ecfr.gov/current/title-47/chapter-I/subchapter-B/part-52/subpart-B/section-52.15) — eCFR / Federal Communications Commission
2. [Advanced Methods To Target and Eliminate Unlawful Robocalls (Reassigned Numbers Database order)](https://www.federalregister.gov/documents/2019/03/26/2019-05620/advanced-methods-to-target-and-eliminate-unlawful-robocalls) — Federal Communications Commission, Federal Register, 2019
3. [Consumer and Governmental Affairs Bureau Announces Interim Usage Charges for the Reassigned Numbers Database](https://www.federalregister.gov/documents/2021/11/05/2021-23855/consumer-and-governmental-affairs-bureau-announces-interim-usage-charges-for-the-reassigned-numbers) — Federal Communications Commission, Federal Register, 2021
4. [Reassigned Numbers Database](https://www.reassigned.us/) — Reassigned Numbers Database administrator, on behalf of the FCC
5. [Messaging Principles and Best Practices](https://api.ctia.org/wp-content/uploads/2023/05/230523-CTIA-Messaging-Principles-and-Best-Practices-FINAL.pdf) — CTIA, 2023

## Frequently asked questions

### How long before a disconnected number is given to someone else?

It depends on the country. In the US, FCC rules let carriers age numbers from residential customers for no less than 45 and no more than 90 days, and from business customers for 45 to 365 days, before reassigning them. Other countries set their own rules.

### How fast does a phone list decay?

There is no reliable universal figure, and published percentages rarely compare. Measure your own: re-check a sample of the list and compare with your last snapshot. The share of numbers whose status changed per month is your decay rate.

### Does a WhatsApp or carrier check tell me a number was reassigned?

Not directly. A change since your last snapshot, such as a different carrier, a new line type or a messenger account that appeared or disappeared, suggests the number may have changed hands. In the US, the Reassigned Numbers Database answers the question for numbers with a known consent date.

### How often should we re-check?

Before any campaign to numbers you haven't contacted recently, and on a regular schedule for lists you keep. A common starting point is monthly for active marketing lists and quarterly for dormant ones, adjusted to the decay you measure.
