How to Read Telecom CDR Reports Properly

A CDR report can tell you in five minutes what a routing issue, payout drop or traffic anomaly might otherwise take days to argue about. For any partner handling premium or revenue-share voice traffic, knowing how to read telecom CDR reports is not an admin task. It is part of margin control.

The problem is that many teams look at CDRs only when there is a dispute. By then, the useful signal is buried under thousands of rows. A better approach is to treat the report as an operating tool - one that shows where calls connected, where they failed, how long they stayed up, and whether the traffic you sent is actually earning what you expected.

What a telecom CDR report is really showing you

A call detail record is a log of what happened to each call attempt. The exact fields vary by platform and carrier, but the purpose is consistent: document the lifecycle of the call for billing, routing analysis, fraud checks and performance review.

For traffic monetisation partners, a CDR report is usually less about the individual caller and more about patterns. You are looking for whether a destination is converting well, whether a number range is underperforming, whether short-duration traffic is rising, and whether billed minutes match operational reality.

A single row rarely tells the full story. Read in volume, however, CDRs show the commercial health of a route.

How to read telecom CDR reports field by field

Start with the timestamp. This tells you when the call attempt entered the network and helps you map changes against campaign launches, routing updates, carrier incidents or daypart performance. If a route looked stable yesterday and weak today, time-based filtering is your first check.

Next is the calling number or source identifier, where available and relevant to your reporting permissions. For B2B teams, this is usually less important than source group, traffic channel or campaign tag. What matters is whether you can attribute call volumes to the traffic source that produced them.

The called number, dialled destination or service number is often where the commercial analysis begins. If you run multiple premium ranges across several countries, this field lets you compare performance by destination, number set or offer.

Then look at the call status or disconnect cause. This is one of the most misunderstood parts of any report. A call can fail for many reasons: user hang-up, no answer, invalid number, network congestion, barred routing, carrier rejection or codec mismatch. If you only separate calls into answered and failed, you miss the reason the route is losing money.

Duration is usually split into total duration and billable duration. That distinction matters. A call may remain active for a certain number of seconds, but only part of that time may be billable depending on answer supervision, billing increments and carrier rules. If you are checking payouts, billable duration is the figure that matters most.

Cost, rate or revenue fields come next. Depending on the platform, you may see buy rate, sell rate, payout rate, revenue share, currency and total earned amount. This is where technical reporting becomes commercial reporting. A route with strong answer rates but poor net return may still be the wrong route to scale.

Finally, check route or carrier identifiers if available. These help when two destinations look similar on the surface but one is underperforming due to a specific upstream supplier.

The key metrics behind the rows

You do not read telecom CDR reports properly by scanning random records. You read them by grouping the data into metrics that affect earnings and route stability.

ASR, or answer-seizure ratio, is the obvious one. It tells you what percentage of call attempts became answered calls. A low ASR can indicate bad traffic quality, routing failures, poor number presentation, destination issues or simple mismatch between offer and audience. There is no universal good ASR because traffic types differ. Audiotext, call centre and IVR traffic all behave differently. The useful question is whether the current ASR is normal for that destination and source.

ACD, or average call duration, is just as important. In revenue-share models, duration often drives return. If ASR holds steady but ACD falls sharply, payout can drop even when volumes look healthy. Short calls are not always a problem - some services are designed that way - but sudden shifts usually deserve attention.

Revenue per call and revenue per minute show whether the route is commercially efficient. Two number ranges may produce similar gross minutes, but one may pay better because of pricing structure, billing increments or stronger retention.

Failed-call ratio and release cause distribution help you see whether losses are operational or traffic-related. If failures are concentrated around network causes, routing may need work. If the majority are caller abandonments after a few seconds, the issue may sit at the source or offer level instead.

What to check first when something looks wrong

When a partner says a route is down, underpaying or behaving strangely, the first useful move is not to inspect every line. Filter by date, destination, number and traffic source, then compare the current period against a known stable period.

Look at call attempts first. If attempts have collapsed, that may be a traffic supply issue rather than a telecom one. If attempts are steady but answered calls have dropped, focus on ASR and release causes. If answered calls are steady but revenue is down, check ACD, billable seconds and rate fields.

That sequence matters because it narrows the fault domain quickly. Too many teams jump straight to payouts, when the real issue started with route quality two steps earlier.

Common mistakes when reading CDR reports

The biggest mistake is treating all answered calls as equal. They are not. A three-second answered call and a three-minute answered call have very different commercial value.

The second mistake is ignoring billing increments. If a destination bills in larger increments, short calls may monetise differently from what raw duration suggests. You need to understand the billing model behind the report, not just the report itself.

Another common error is mixing time zones. If your media buying team, supplier and reporting portal all use different time references, a same-day comparison can become misleading very quickly.

There is also the issue of sample size. Reading too much into a few dozen calls is risky, especially on new destinations. Trends become useful when the traffic base is large enough to smooth out normal variance.

Finally, many operators focus only on totals. Totals are useful for finance, but performance decisions usually come from segmentation by destination, source, number, hour and release cause.

How to use CDRs for better routing and payout decisions

Good CDR analysis is not just defensive. It helps you move traffic where it earns better and where reporting stays clean.

If one destination shows stable ASR, healthy ACD and consistent release causes, that route may justify more traffic. If another destination has attractive headline rates but weak answer performance, the real yield may be worse than it looks on paper.

This is where live reporting matters. A platform with real-time or near real-time CDR visibility lets you react before a full day of poor traffic accumulates. You can pause a source, test a number range, switch traffic distribution or raise a routing query while the evidence is still fresh. For partners working across multiple international destinations, that speed is often the difference between preserving margin and explaining loss after the fact.

If you use a self-service portal, build a routine around it. Check daily for major swings, weekly for route trends and monthly for payout reconciliation. A report is only useful if you read it before the commercial damage is done.

A simple workflow for reading telecom CDR reports

Start broad, then narrow down. First review total attempts, answered calls, ASR, ACD and revenue for the period. Then segment by destination and number range. After that, isolate any route with an abnormal change and inspect call statuses, release causes and billable duration.

If the issue still is not clear, compare by hour and by source. A route that performs badly only during a certain window may point to congestion or a supplier-side issue. A route that performs badly only from one source may point to traffic quality.

Most importantly, record what normal looks like for each route. Without a baseline, every fluctuation feels urgent. With a baseline, you can tell whether a route is genuinely degrading or simply behaving as expected.

Teams that do this well are usually not the teams with the fanciest dashboards. They are the teams that understand which numbers matter, question anomalies early and reconcile technical data with payout outcomes. That discipline turns a CDR report from a spreadsheet into a decision tool.

The best time to inspect your reports is before you need to challenge a result. Once you can read the pattern behind the rows, you are not just checking calls - you are protecting route quality, partner trust and revenue at the same time.