Marketplaces Reduce Seller Churn by Proving ROI With Call Data

Sellers churn from marketplaces for two reasons. The first is that the marketplace genuinely isn’t working for them, the leads aren’t there, the buyer intent is low, the listing quality is poor. The second, far more common reason is that the marketplace is working, but the seller can’t see it clearly enough to believe it.

Call data closes that gap. A seller who can see how many qualified buyers called their listings, how many of those calls reached a human, and how many resulted in a confirmed appointment has a fundamentally different view of what their subscription is producing than a seller who only sees traffic and click numbers.

The retention argument is not just that call data is more convincing evidence. It is that call data is the right evidence, because the conversation the seller is trying to have “are my leads actually good?” is a question that impression reports structurally cannot answer.

Why sellers churn even when a marketplace is actually working for them

The story sellers tell when they churn is almost always about lead quality. The leads aren’t converting. The callers aren’t serious. The traffic doesn’t translate. What they are usually describing, without realising it, is a visibility problem.

Sellers are often complaining about lead quality as opposed to lead volume, particularly for horizontal marketplaces.

“It’s a negotiation tactic for dealers when they’re fighting marketplaces who are increasing their pricing year on year” says Mayu Fenton-James, Enterprise Customer Success Manager at Kaisa.

A seller who receives 80 inbound calls in a month and converts 12 of them to appointments has, objectively, a 15% appointment rate. But if that seller’s only reporting shows 80 calls received,they have no way to evaluate whether 12 appointments is good, bad, or par for the platform. They cannot see their own qualification rate. They cannot compare their conversion rate to peer dealers. They have a volume number and a gut feeling  and gut feelings, especially after a quiet week, tend toward pessimism.

“The age-old debate for marketplaces has been to say: look how many leads we create for you. And the dealer says, most of them are poor quality anyway,” says Henrik Lenerius, Chief Product Officer at Kaisa. “Now you have a much stronger evidence base to say, actually, this was valuable to you. It’s quality over quantity, proving your return on investment to your customers.”

Churn, in most cases, is not a product problem. It is a proof problem. The value is being delivered. The evidence is not.

What call data shows that click and impression reports don’t

Click and impression data answers one question: did buyers engage with this listing? It says nothing about what happened after the click — whether the buyer called, whether someone answered, whether the enquiry concerned a real, available vehicle, or whether the conversation progressed toward a sale.

Call outcome data answers the questions sellers are actually asking at renewal time.

Who called? Call analytics identifies whether callers were genuine buyers, enquiring about available stock, asking about price, requesting a test drive, or whether the call volume included misdialed numbers, spare-parts enquiries, and callers who spent 90 seconds in an IVR before hanging up. A seller who can see their qualification rate understands what their leads actually looked like, not just how many arrived.

Did anyone answer? Connection rate the share of calls that actually reached a human  is often the most surprising number a seller sees for the first time.

“On paper, the dealer answered every lead. In reality, a lot of them hit voicemail,” says Mayu Fenton-James. Call analytics separates received from connected, which is often where the real conversation about staffing, after-hours coverage, and missed revenue begins.

What was the outcome? Appointment detection, whether a test drive or showroom visit was agreed on the call, is the outcome metric closest to revenue. A seller who can see that 12 of their 80 inbound calls resulted in a confirmed appointment is looking at a number that is the closest indicator they have to their monthly sales pipeline. That is the number that changes the renewal conversation.

James Morris CX a and data specialist at Kaisa makes a related point: the value compounds once the data connects end to end, from the first call through to a confirmed appointment, rather than sitting as separate, disconnected numbers.

Building call data into the renewal conversation

The renewal conversation typically starts with the seller’s question: is this working? The account manager’s job is to answer it with evidence the seller trusts.

Traffic and impression data rarely settles that question. Sellers have learned to be sceptical of reach metrics — too many of them have seen high click counts and low sales to accept one as proof of the other. What sellers trust is a conversation that sounds like their own business.

“Sellers can claim they’re not getting good leads at renewal meetings, and if the marketplace has no data or visibility to fight that argument what can they do?” adds Mayu Fenton-James.

A practical renewal framework using call data looks like this:

Start with connection rate. Show the seller what share of their inbound calls actually reached a human. If the rate is lower than expected, this opens a conversation about how to close it — staffing, hours, IVR configuration — rather than a debate about whether the marketplace is generating leads.

Connect outcomes to the listing. Where call analytics links outcomes back to specific vehicles, the conversation becomes even more specific: which listings are generating qualified calls, which are attracting misdials or dead-end enquiries, and where a listing adjustment might improve conversion. The marketplace is no longer just a traffic source. It is a diagnostic tool.

Spotting at-risk sellers before they churn

The most valuable use of call data in retention is not at renewal time. It is the 60 to 90 days before renewal, when the seller’s sentiment is forming but the decision has not yet been made.

Certain patterns in call data are reliable early-warning signals. A seller whose call volume drops sharply, without a corresponding drop in listing count, may have experienced a change in listing placement or algorithm weighting worth investigating.

“There’s far more buyers getting stuck in IVR loops and voicemails than marketplaces realise – a lot of callers never actually end up interacting with a human,” Fenton-James notes. “That really shows marketplaces that dealers might not be handling leads as well as they thought. A dealer might say I answered all these leads, when actually the lead was hitting a voicemail.” A seller whose qualification rate declines over consecutive months is having fewer real buyer conversations, which is typically the pattern that precedes a “the leads aren’t converting” churn conversation.

A seller who stops logging into their call analytics dashboard entirely is a different kind of signal: disengagement from the data often precedes disengagement from the platform.

None of these signals requires a manual audit. A reporting system that surfaces declining qualification rates or connection rates by seller account gives customer success teams a prioritised list of accounts to check in with, proactively, before the renewal conversation arrives with a seller who has already made up their mind.

The intervention at 60 days out is a very different conversation to the one at renewal. It is a “we noticed this and wanted to talk” conversation rather than a “please don’t leave” one. The outcome rates are correspondingly different.

What this requires from your reporting tools

Not every call analytics implementation produces the reporting that customer success teams need for this use case. Three things specifically matter.

“Seller-facing visibility into the numbers, not just the account manager’s. An account manager presenting call data to a seller in a review meeting is useful. A seller who can see the same numbers between meetings, through whatever reporting the marketplace already gives them, is significantly more engaged, and significantly less likely to develop an unexamined sense that the platform is not working. Most marketplaces do this by pushing call outcome data into their existing seller-facing reporting, rather than setting  up a separate dashboard login. Ongoing visibility is what matters, not a specific interface.”

Exportable, presentation-ready reports. Customer success teams should be able to pull a clean summary of a seller’s call performance, connection rate, qualification rate, appointment rate, trend over time  without manual data extraction. If producing the report requires a data team or a custom export, it will not happen consistently across the account base.

Network benchmarks the account manager can show. A seller’s own numbers are useful. A seller’s numbers alongside what top-performing peers on the same platform are achieving is a conversation. Benchmarks require consistent data collection methodology across all sellers,  which is the same requirement that makes platform-wide call intelligence possible in the first place.