How to Monitor Call Quality on Ringba

You're running campaigns on Ringba. The calls are flowing. Publishers are getting paid. Buyers are receiving transfers. From a routing perspective, everything works.

But here's the question nobody on your team can answer with confidence: what's actually happening on those calls?

Ringba gives you excellent call tracking and routing data — duration, connection status, which publisher sent it, which buyer received it. What it doesn't tell you is whether the caller was being coached, whether the lead was genuinely qualified, or whether a publisher's traffic has a compliance problem that's quietly building into a liability.

That's the gap Ringba call quality monitoring fills. And if you're running any meaningful volume, it's a gap you can't afford to leave open.

Why Call Tracking Isn't Call Quality

Ringba tracks what happened to a call. QC tells you what happened on a call.

Ringba knows that a call from Publisher X lasted 4 minutes and 12 seconds, connected to Buyer Y's target, and met the duration threshold. That's useful routing and billing data. What Ringba doesn't know is whether the caller fabricated their situation, whether the agent handled compliance correctly, or whether the "qualified" lead will ever convert.

Those answers live in the recording. And at scale, nobody is listening to the recordings.

A QC analyst can review 15-20 calls per hour. If you're running 500 calls a day through Ringba, full manual review would require a team of four or five people working full shifts. So you sample — maybe 30 calls a day across all publishers. That means 94% of your traffic goes unheard. Fraud, compliance issues, and quality problems hide in that unreviewed majority.

Automated Ringba call quality monitoring closes that gap. Every call gets transcribed and analyzed. Flags show up the same day. And critically, every result is tied back to the publisher who sent it — so you can see quality metrics by source, not just in aggregate.

How Ringba QC Integration Works

Ringba uses a per-campaign tracking pixel system. Unlike platforms that support a single global webhook for all campaigns, Ringba requires you to add a pixel URL to each campaign individually. This means a few extra minutes of setup per campaign, but the result is the same: every call that matches your trigger conditions gets sent for QC analysis.

Here's the flow:

1. A call ends on a Ringba campaign and a recording becomes available.
2. The pixel fires, sending call metadata (caller number, duration, recording URL, publisher info) to your QC endpoint via a GET request.
3. The recording is transcribed and analyzed asynchronously — typically within minutes.
4. Results appear in your QC dashboard: full transcript, AI-generated summary, disposition classification, and any detected red flags.

The flags that matter for pay-per-call QC:

Flag	What It Means
Coached Call	Caller is being fed fabricated information or scripted responses to fraudulently qualify
Compliance Issue	Caller doesn't match campaign qualifiers — wrong state, demographic, or misrepresentation
DNC Violation	Potential Do Not Call list violation
TCPA Violation	Potential Telephone Consumer Protection Act issue

Each call is analyzed independently based on its own transcript. Processing is asynchronous — results appear within minutes of the call ending, not during the call itself. For QC purposes, same-day detection is what matters. Catching a problem today is infinitely better than discovering it when the buyer calls to complain next week.

Setting Up Per-Campaign QC on Ringba

The setup uses Ringba's built-in tracking pixel feature. You'll create a pixel URL that passes call data as query parameters, then add it to each campaign you want monitored.

Step 1: Get Your API Key

Sign up at ConvoQC and grab your API key from the integrations page. You get $10 in free credit — enough to process roughly 10 hours of call recordings at $0.015/minute.

Step 2: Build Your Pixel URL

Ringba pixels use GET requests with token placeholders in the query string. When the pixel fires, Ringba replaces each token with the actual call data.

Here's the full pixel URL with all relevant Ringba tokens:

https://dash.convoqc.com/api/analyze?api_key=YOUR_API_KEY&caller_number=[Call:InboundPhoneNumber]&duration_seconds=[tag:CallLength:Total]&recording_url=[tag:Recording:RecordingUrl]&ringba_call_id=[Call:InboundCallId]&campaign_name=[tag:Campaign:Name]&call_timestamp=[Call:CallConnectionTime]&publisher_id=[tag:Publisher:Id]&publisher_name=[tag:Publisher:Name]&sub_id=[tag:Publisher:SubId]&buyer_id=[tag:Buyer:Id]&buyer_name=[tag:Buyer:Name]

Step 3: Understand What Each Token Sends

Ringba Token	What It Sends	Why It Matters
[Call:InboundPhoneNumber]	Caller's phone number (E.164)	Identifies the caller, cross-references with DNC indicators
[tag:CallLength:Total]	Call duration in seconds	Contextualizes flags — a 30-second flag is different from a 5-minute one
[tag:Recording:RecordingUrl]	URL to the call recording	The actual audio that gets transcribed and analyzed
[Call:InboundCallId]	Ringba's unique call ID (RGB prefix)	Links the QC result back to the specific call in Ringba
[tag:Campaign:Name]	Campaign name	Groups results by campaign for reporting
[Call:CallConnectionTime]	When the call connected	Ties QC results to your timeline
[tag:Publisher:Id]	Publisher's ID in Ringba	Identifies the traffic source for quality tracking
[tag:Publisher:Name]	Publisher display name	Human-readable source name in reports
[tag:Publisher:SubId]	Sub-affiliate tracking ID	Drills down to the specific sub-source within a publisher
[tag:Buyer:Id]	Buyer's ID in Ringba	Tracks which buyer received the call
[tag:Buyer:Name]	Buyer display name	Buyer-specific quality reporting

The publisher tokens — [tag:Publisher:Name], [tag:Publisher:Id], and [tag:Publisher:SubId] — are the most important for QC purposes. They're what connect every flag, every disposition, and every quality metric back to the traffic source that sent the call. Without them, you'd know 5% of your calls have problems but not which publisher is responsible.

Step 4: Add the Pixel to Your Campaign

In Ringba, navigate to Campaigns, select the campaign you want to monitor, and go to the Tracking Pixels section. Add a new pixel and paste your URL. Set it to fire on the Recording event — this ensures the recording URL is available when the pixel triggers.

Step 5: Repeat for Each Campaign

This is the main difference from platforms with global webhooks. On Ringba, each campaign needs its own pixel. If you're running 10 campaigns, that's 10 pixel configurations. The URL is identical for every campaign (same API key, same tokens) — you're just adding it in multiple places.

The important operational note: when you launch a new campaign on Ringba, add the QC pixel as part of your launch checklist. If you forget, those calls won't get analyzed. There's no global fallback catching them.

A Note on Ringba Terminology

If you're coming from another platform, Ringba's naming conventions can trip you up:

• Publisher = the traffic source or affiliate sending calls. This maps directly to what you'd call a publisher or traffic source on any platform.
• Buyer = the company purchasing leads. This is who pays for qualified calls.
• Target = a specific routing endpoint under a Buyer. Think of it as an individual phone number, agent, or call center queue that receives routed calls.

When you see publisher-level QC data in your dashboard, it corresponds to [tag:Publisher:Name] from Ringba. When you drill into sub-affiliate performance, that's [tag:Publisher:SubId].

What to Look for in Your Ringba Publisher Traffic

Once calls are flowing through QC analysis, here's what to watch — and what the data actually tells you.

Flag Rates by Publisher

This is the single most actionable metric. Pull up flag rates broken down by publisher and look for outliers. If your network average flag rate is 3% and one publisher is running at 12%, that's not noise — that's a quality problem.

Pay particular attention to coached call flags. A publisher with an elevated coached call rate is either running poor traffic sources or actively gaming your payouts. Either way, it demands a conversation. Having the specific data — "14 of your 180 calls this week were flagged for coaching indicators" — turns a vague quality concern into an actionable publisher conversation.

Sub-ID Performance

Ringba's [tag:Publisher:SubId] token gives you visibility one level deeper than the publisher. A publisher might have an overall acceptable flag rate, but one specific sub-affiliate could be the source of all the problems. Sub-ID data lets you tell a publisher exactly where the issue is, rather than making a blanket accusation about their traffic.

Duration vs. Flag Correlation

Short calls with flags tell a different story than long calls with flags. A 30-second flagged call is probably a wrong number or an obviously unqualified caller — annoying but low-cost. A 6-minute flagged call where the caller sounded qualified but was actually coached — that's the expensive one, because the buyer invested real agent time.

Sort your flagged calls by duration. The longer ones deserve priority review.

Campaign-Level Patterns

Since Ringba pixels are per-campaign, you can see which campaigns attract more quality issues. Some verticals naturally have higher flag rates than others. If your Medicare campaign consistently shows elevated coached call flags while your home services campaign runs clean, the issue might be vertical-specific rather than publisher-specific. Adjust your thresholds accordingly.

The Per-Campaign Trade-Off

Ringba's per-campaign pixel model has a genuine disadvantage compared to platforms with global webhooks: you can miss campaigns. If your team launches a new campaign and skips the pixel, those calls go unmonitored. At scale, this happens more often than you'd think.

Build it into your process. Add "configure QC pixel" to your campaign launch checklist, right next to "set routing rules" and "configure payouts." Periodically audit your active campaigns against your QC coverage to catch any gaps.

The upside of the per-campaign model is granularity. You can choose to monitor only specific campaigns — useful if you're testing the QC process on your highest-volume campaigns first before rolling out to everything.

What Doesn't Get Caught

Automated QC analyzes each call's transcript for red flags. There are things it doesn't do, and setting accurate expectations matters.

It doesn't do cross-call pattern analysis. If the same caller phones in through three different publishers in a week, each call is evaluated independently. Detecting that pattern requires separate analysis of caller numbers across your volume.

It doesn't scrub against the DNC registry in real time. A DNC flag means something in the conversation indicated a potential issue — the caller mentioned being on a list, or the context suggests a compliance concern. It's not a registry lookup.

Processing is asynchronous, not real-time. Results appear within minutes of the call ending. You won't get a flag while the call is still active. For QC purposes, this is fine — you're building publisher quality data over time, not making in-call routing decisions.

Turning QC Data Into Publisher Decisions

The real value isn't in individual call flags. It's in what the aggregate data tells you about your publishers over time.

When every call from every publisher gets analyzed, you can calculate objective quality metrics by source: flag rate, qualification rate, average duration, disposition breakdown. You stop guessing which publishers are good and which are costing you money. You start ranking them on data.

Set internal thresholds. Any publisher above a 5% flag rate gets a documented review. Above 10% gets volume restrictions. Above 15% gets paused pending investigation. The specific numbers depend on your verticals and tolerance, but having thresholds at all puts you ahead of most operations.

Reward your best publishers too. A publisher maintaining a 1% flag rate and a 45% qualification rate over 500 calls is sending genuinely valuable traffic. That's a relationship worth protecting with priority routing and premium payouts.

Getting Started

If you're running Ringba campaigns without systematic QC, here's the practical path. Pick your two or three highest-volume campaigns and add the pixel. Let it run for a week. Look at the data — particularly flag rates by publisher and the transcripts of flagged calls. That week of data will tell you more about your traffic quality than months of occasional spot-checking.

ConvoQC integrates with Ringba through the per-campaign pixel setup described above, processing calls at $0.015 per minute of audio with $10 in free credit on signup. The setup takes about 5 minutes per campaign, and results start appearing as soon as calls come in.

The calls are already being recorded. The data is sitting there. The only question is whether you're going to look at it before or after it becomes a problem.