See the numberthey see
Place a real test call and read back the caller ID the handset actually displayed, so you know when a route rewrites it.
A route sold as CLI can still rewrite, truncate or strip the number somewhere down the chain, and you usually find out when answer rates fall.
Caller ID testing places genuine calls to handsets in the destination country and reports what was actually displayed. Qualify a route before a campaign goes on it, check every member of a Revenue Turbine blend, and hold a supplier to the CLI they sold you with a dated result.
CLI preservation checks
Confirm the full number survives the route.
Per-route testing
Test before you commit traffic to a route.
Blend member testing
One test per member, so you can see which supplier alters the number.
Result history
A record of every test, so you can prove degradation over time.
Results as they land
A person with a handset in the destination country answers and reports, so allow up to 24 hours.
One real call, one clear answer
A test travels over the same route your traffic will, so the answer describes that route and not a lab.
- 01
Choose what to test
A marketplace route before you buy it, a route you own or bought, every member of a Revenue Turbine blend, or the SIP address of a switch we do not host.
- 02
Set the caller ID
Enter the number to present in international format. The destination is fixed to the route's country, and you see whether a handset there is reachable now, and what the test costs, before you request it.
- 03
A real call goes out
The call is placed over the route to an opted-in handset in that country, presenting your caller ID. The person holding the phone reports the number that displayed.
- 04
Read the verdict
Match, Partial, Replaced or Restricted, with the number you sent beside the number shown and the time it completed. You hear about it in the dashboard and by email.
The number you sent, beside the number that arrived
Every completed test comes down to two numbers and a verdict. Match means your caller ID displayed exactly as sent, with +44 and 0044 treated as the same number. Partial means a truncated or reformatted version arrived. Replaced means the network overwrote it with a different number, and Restricted means nothing displayed at all. Each test is timestamped from request to report, so the conversation with a supplier starts from evidence rather than impressions.
- A real person on a real handset reports what displayed
- Created, sent to tester and completed times recorded on every test
- A plain explanation and a next step alongside every verdict
- Results in the dashboard, by email and through the API
The destination network overwrote your caller ID before it displayed. The handset showed a different number.
Try a different route, or a caller ID local to the destination.
- Route
- UK Mobile, Standard tier
- Test destination
- United Kingdom
- Repeat
- Weekly
- Created09:12
- Sent to tester09:14
- Completed09:41
Find the member that changes your number
A blended route is only as good as the member carrying a given call. Test the blend and a real call is placed through each member route, every one presenting the same caller ID into the same destination, with the results lined up side by side. One weak supplier can no longer hide behind a good average, and you can rebalance the blend before a buyer notices anything.
- One test per enabled member route, each placed and reported on its own
- Results grouped under the blend and compared member by member
- The same caller ID and destination on every leg, so the comparison is fair
- Route 1Direct+44 7700 900123Match
- Route 2Premium7700 900123Partial
- Route 3Standard+44 7700 900555Replaced
- Route 4StandardWithheldRestricted
One real call per member route, each placed and reported as its own test.
Catch the week a route changes
A route that delivered your caller ID last month can quietly start rewriting it. Schedule a test for a set time or repeat it daily or weekly, and every run is a fresh call to a real handset, booked as soon as the previous result is in. The history keeps each number sent, each number shown and when, so you can point to the exact week a supplier's caller ID handling slipped.
- Run now, at a time you choose, daily or weekly
- Cancel any test before it runs, at no charge
- Filter the history by route, by blend or by status
- 14 Sep14 Sep 2026+44 7700 900987Replaced
- 7 Sep7 Sep 20267700 900123Partial
- 31 Aug31 Aug 2026+44 7700 900123Match
- 24 Aug24 Aug 2026+44 7700 900123Match
- 17 Aug17 Aug 2026+44 7700 900123Match
Built so the result can be trusted
Caller ID testing is only worth running if the answer is real and the process is fair to the route, so both are designed in from the first step.
Real handsets, real people
Every test ends on an opted-in handset in the destination country, and the person holding it reports what displayed. Nothing about the result is simulated or inferred.
Test before you buy
Any public voice route on the marketplace can be tested before you purchase it. Private routes can be tested by their owner or by anyone holding an active purchase.
Switches we do not host
Give us a SIP address, an optional port, the caller ID and the handset country, and we test it, with no listing required. Your switch only needs to accept calls from our edge addresses.
Every member of a blend
A blend test places one call through each enabled member route, so you can see exactly which supplier delivers your number and which one alters it.
Knows where it can test
The test form shows whether a handset is reachable in the destination right now. A destination with nobody available is refused before a test is created, so nothing is charged.
Scheduled and recurring
Run a test straight away, at a time you pick, or daily or weekly. Each repeat is a new call to a real handset, booked once the previous result has come back.
Results that find you
Every outcome arrives as an in-app notification and an email, and the full history of tests, verdicts and numbers shown is in the dashboard and available by API.
Fair to the route
If no handset can take the call, the test closes as not tested: no verdict against the route and no charge. It is never recorded as the route failing.
Stop paying for a caller ID you are not getting
Caller ID problems rarely announce themselves. They surface weeks later as falling answer rates. A test turns a suspicion into a dated fact, on the same account you buy your routes with.
Protect the answer rate
People answer numbers they recognise. Confirm the local or branded caller ID you are paying for really arrives before a campaign spends a single minute against it.
Evidence, not anecdotes
Each result pairs the number sent with the number shown and the time it was reported, which is exactly what a supplier conversation, a credit request or a route issue needs.
Wired into the marketplace
Test a route before you buy it, test the members of your own blends, and keep every result on the account that already holds your routes and your balance.
Charged only when a test can run
A destination with no handset online is refused up front, a cancelled test costs nothing, and a test that never reaches a handset closes without a verdict or a charge.
Caller ID checks, done properly
How caller ID usually gets checked, and what a test on the exchange changes.
| The usual way | On PacketExchange | |
|---|---|---|
| How it is checked | Someone rings their own mobile from the office and hopes it is representative. | A real call over the route itself, to a handset in the destination country. |
| When you find out | When answer rates fall and someone starts investigating. | Before traffic goes on the route, and on a daily or weekly schedule after that. |
| Blends | Each supplier checked separately by hand, if at all. | One request tests every member route and lines the results up side by side. |
| Before buying | Buy the route, send traffic and find out later. | Public routes can be tested before you buy them. |
| Proof | A screenshot and a description of what someone saw. | A dated record of the number sent and the number shown. |
| Inconclusive tests | Reported as a failure, as though the route were broken. | Closed as not tested, with no verdict on the route and no charge when no handset was reached. |
Protect answer rates
A number that arrives mangled or withheld gets ignored.
Qualify a supplier
Check the CLI claim before you commit volume.
Settle a dispute
Show exactly what arrived, and when.
A flat price per test, shown before you run it
Each test has one flat price, shown in the test form alongside how many tests you have left this hour, and it is debited from your balance only when the test runs. A blend test counts as one test per member route.
There is no subscription. Nothing is charged when a destination has no handset online, when you cancel a test before it runs, or when a test closes because no handset could be reached.
Browse routes to testFlat per test
One price per test call, visible in the form before you request it.
From your balance
Debited when the test runs, from the same prepaid balance as your traffic.
Blends
One test per enabled member route, each priced as a single test.
Hourly allowance
A rolling hourly limit per account, with the tests remaining shown as you book.
No subscription, no minimum, and no charge for a test that never reaches a handset.
Caller ID testing, answered
Who answers the test call?
A person with an opted-in handset in the destination country. The call arrives on their phone like any other, and they report the caller ID that displayed. That report is the result.
How long does a result take?
Allow up to 24 hours, because a person has to answer and report. You are notified in the dashboard and by email as soon as the result arrives, and the test page refreshes itself while you wait.
What do Match, Partial, Replaced and Restricted mean?
Match: your caller ID displayed exactly as sent. Partial: a truncated or reformatted version displayed. Replaced: the network overwrote it with a different number. Restricted: the caller ID was withheld and nothing displayed.
What if nobody in the destination can take the call?
If no handset is online when you ask, the test is refused before it is created. If none comes online within 24 hours, the test closes as not tested with no charge. If the call was placed but the handset owner never reported, it also closes as not tested, and the test is charged because the call was made.
Can I test a route before I buy it?
Yes. Any public voice route on the marketplace can be tested before purchase. Private routes can be tested by their owner and by buyers with an active purchase.
Can I test a switch that is not on PacketExchange?
Yes. Enter the SIP address and port of the switch, the caller ID to present and the handset country. We create a private, unlisted test target and run the ordinary test. Your switch needs to accept calls from our edge addresses.
Which country is the handset in?
For a route, the destination country of that route, because caller ID display is specific to the country. For a switch test, you choose from the countries that have a handset online at that moment.
Can tests run automatically?
Yes. Schedule a test for a set time, or repeat it daily or weekly. Each repeat is a new call to a real handset, booked once the previous result is back, and any test that has not run yet can be cancelled.