Automotive Marketing

AI Sales Call Analysis for Canadian Auto Dealers: Conversica vs Foureyes vs Cars.com

15+ years in local marketing; Google Ads certified; Shopify Partner.

You hired a BDC rep six months ago. She's logging calls in the CRM. The manager says the team is following up. But your close rate on internet leads is sitting at 9%, and you can't figure out why.

Here's the thing: the answer is almost certainly sitting in your call recordings. You just don't have time to listen to 200 calls a month. That's exactly what ai sales call analysis tools are supposed to fix. They listen for you, flag the problems, and tell you where deals are dying.

But "supposed to" is doing a lot of work in that sentence. This article breaks down three of the main tools Canadian dealers are evaluating right now: Conversica, Foureyes, and Cars.com Conversations. What each one actually does, what it costs, where it falls short, and how to decide which one fits your store. I'll also flag the compliance wrinkles that matter specifically in Canada, because most of these vendors were built for the U.S. market and some of them show it.

This is not a guide to your overall dealership digital marketing strategy. For that, see our complete auto dealership marketing guide. This article is specifically about AI call analysis tools and how to evaluate them.

What AI Sales Call Analysis Actually Does (and Doesn't Do)

Let's start with the basics, because the category name gets stretched in a lot of directions by vendors.

Real AI call analysis does three things. It transcribes your calls automatically. It scores those calls against a set of criteria (did the rep ask for the appointment? did they get the customer's name? did they overcome the price objection?). And it surfaces patterns, so you can see that your Monday morning team closes at 14% but your afternoon team closes at 6%, and then actually go find out why.

What it doesn't do: it doesn't replace a human sales manager who knows your market. It doesn't tell you whether the deal was profitable. And it definitely doesn't fix a broken follow-up process on its own.

I think a lot of dealers get sold on the idea that the tool is the fix. It's not. The tool shows you where the problem is. You still have to fix it.

Per the 2024 CADA CARTS study of 549 Canadian dealers, 59% of dealerships report underutilizing the technology they already have. Call analysis tools fall squarely into that category. Dealers subscribe, get excited for 30 days, and then nobody's reviewing the reports.

So before you spend money on any of these tools, ask yourself: who on my team will actually look at this data every week? If the answer is "nobody right now," that's the thing to fix first.

Conversica: AI That Follows Up, Not Just Listens

Conversica is a bit of a different animal from the other two tools in this comparison. It's not primarily a call recording analysis platform. It's an AI-powered follow-up assistant that reaches out to leads via email and SMS, has a back-and-forth conversation with the prospect, and then hands off to a human when the lead is ready to talk.

The call analysis piece comes in as a secondary feature. Conversica can review inbound calls and flag whether a lead was properly followed up, whether the AI assistant's outreach matched the conversation, and whether there are gaps in the handoff process.

Where it's strong: Conversica is genuinely good at the follow-up problem. If your BDC is dropping leads after the first call, Conversica's AI will keep working that lead for weeks without your team having to do anything. That's real.

Where it gets complicated for Canadian dealers: Conversica's outbound AI messaging runs into CASL (Canada's Anti-Spam Legislation) in a way that U.S.-focused vendors often don't account for. CASL requires express or implied consent before sending commercial electronic messages. Conquest leads, aged leads, and third-party list leads are high-risk territory. If Conversica is blasting follow-up emails to leads where you don't have clear consent documentation, you're exposed.

I've seen dealers get excited about the automation and not ask the CASL question. Ask the CASL question.

Pricing: Conversica doesn't publish pricing publicly. Based on what dealers report in forums, expect somewhere in the $1,500–$3,000 USD/month range for a single-rooftop dealership, though multi-rooftop groups negotiate differently. Convert that to CAD and you're at roughly $2,000–$4,000/month depending on the exchange rate. That's a real budget line.

OEM co-op eligibility: Not confirmed as an OEM-approved vendor for major Canadian OEM programs as of early 2026. Verify with your zone rep before assuming co-op dollars can cover this.

Foureyes: The Call Tracking and Visitor Intelligence Play

Foureyes is more squarely in the call analysis and visitor intelligence category. The core product tracks who's visiting your website, ties those visits to phone calls, and then uses AI to score those calls and surface missed opportunities.

The specific feature that gets dealers' attention is the "missed opportunity" alert: Foureyes can identify when a customer visited your VDP (vehicle display page), called in, and then wasn't followed up with properly. It can also flag when a customer who's been on your site multiple times hasn't been contacted at all.

What the AI analysis actually looks at: Foureyes scores calls on whether the rep asked for an appointment, whether they captured contact information, whether they handled the price question, and whether the call ended with a clear next step. It's not magic, but it's consistent in a way that a human manager reviewing calls ad hoc usually isn't.

The inventory feed wrinkle: Here's something most vendors won't tell you upfront. If your website runs on Dealer.com, DealerOn, CDK, or Dealer Inspire, Foureyes has to integrate with those platforms to tie VDP visits to call data. Some of those integrations are cleaner than others. Dealer.com in particular has historically been protective about third-party data access. Before you sign with Foureyes, get a straight answer from your website provider about what data Foureyes can actually pull. If the integration is partial, your "missed opportunity" reports will have gaps.

Where it's genuinely useful: In my experience, dealerships that use Foureyes and actually review the weekly missed-opportunity report with their BDC manager see real changes in follow-up behaviour within 60 days. Not because the AI is magic, but because making the missed calls visible creates accountability.

Pricing: Foureyes is generally in the $500–$1,500/month range for a single rooftop, depending on the package. More accessible than Conversica for a smaller store.

Canadian compliance note: Foureyes uses call recording, which in Canada requires one-party or two-party consent depending on the province. In Ontario, one-party consent is generally sufficient. In B.C., the situation is more nuanced under PIPA (Personal Information Protection Act). Make sure your call recording disclosure is in your phone greeting. "This call may be recorded for quality assurance" isn't just a courtesy; in some provinces it's a legal requirement.

Cars.com Conversations: The Per-Lead Marketplace Problem

Cars.com Conversations is a different product category again. It's primarily a managed chat and messaging tool that Cars.com bundles with its marketplace listings. The "AI" component handles initial chat responses on your VDPs hosted on Cars.com, and the conversation analysis piece lets you review how those chats are going.

Here's the thing about Cars.com Conversations specifically: it's hard to evaluate it in isolation from the Cars.com marketplace itself. And that's where I'd push back on most dealers who are considering it.

Per the research, dealers are increasingly frustrated with per-lead marketplaces. The math that keeps coming up: paying $400 per lead on deals that gross $800 isn't a marketing budget, it's a subsidy. Cars.com, CarGurus, TrueCar, and Autotrader all operate on models that erode your front-end margin. The AI chat analysis tool is a feature layered on top of that underlying cost structure.

What the AI analysis actually gives you: Insight into how the chats initiated from Cars.com listings are converting, whether leads are being followed up, and where conversations are going cold. That's useful data. But it's data about leads you're already paying a premium for.

The honest assessment: If you're already spending significantly on Cars.com and you're trying to squeeze more value out of that spend, the Conversations tool might help you convert more of the leads you're already buying. But it doesn't solve the underlying problem, which is the cost per acquisition on marketplace leads.

For a full breakdown of how PPC and first-party lead generation compare to marketplace lead costs, see our dealership PPC and Google Ads strategy guide.

OEM co-op: Cars.com is generally an approved vendor in most major OEM co-op programs. That matters if you're trying to run co-op dollars through this.

What to Actually Measure When You're Evaluating These Tools

This is the section most vendor pitches skip. They show you the dashboard. They show you the heat maps. They show you the AI scoring interface. What they don't show you is how to know if the tool is actually changing your numbers.

Here's a worked example. Say your dealership takes 400 inbound calls a month. Your BDC closes 9% of those into appointments, and 40% of appointments show. That's roughly 14 shown appointments from 400 calls. If your close rate on shown appointments is 60%, you're closing about 9 deals a month from inbound calls.

Now say an AI call analysis tool identifies that your team is failing to ask for the appointment on 35% of calls. You fix that. Your appointment rate goes from 9% to 12%. That's 48 appointments instead of 36, 19 shows instead of 14, and at 60% close: 11-12 deals instead of 9. On a $2,500 front-end gross, that's $5,000–$7,500 in additional front-end per month.

That math is why these tools can pay for themselves. But only if someone is actually reviewing the call scores, running the training, and holding the BDC accountable. The tool surfaces the problem. You still have to fix it.

The metrics to track before you start, so you can measure after:

Inbound call volume per month (pulled from your phone system)
Appointment set rate from calls (from your CRM)
Show rate on those appointments
Close rate on shows
Average front-end gross per deal from inbound call leads

Get those four numbers before you sign anything. If you can't pull them, that's the first thing to fix.

How a Real Evaluation Should Go (Week by Week)

Most dealers evaluate these tools by watching a demo and signing a contract. That's backwards. Here's how I'd actually do it.

Week 1: Baseline your current call data. Pull 30 days of call recordings from your existing phone system. If you don't have call recording, that's the first gap. Most modern phone systems (Mitel, RingCentral, Dealertrack's phone integration) have this built in. Have your sales manager or BDC manager manually review 20 calls and score them on a simple rubric: did the rep get the name, ask for the appointment, handle the price question, confirm next steps? Write those scores down. That's your baseline.

Week 2: Request trials from the vendors you're considering. Conversica, Foureyes, and Cars.com Conversations all offer demos and some offer limited trials. During the demo, ask specifically: "Show me a call scoring report from a dealer similar to mine." Not a mock-up. An actual report. If they can't show you a real report, that tells you something.

Week 3: Run the integration checklist. Before you commit to anything, get your website provider (Dealer.com, DealerOn, CDK, whoever) on a call with the vendor's implementation team. Ask directly: what data can this tool access on our platform? What requires a workaround? What can't it access at all? Document the answers. A tool that can't see your VDP traffic because of a platform restriction is missing half its value proposition.

Week 4: Check the compliance boxes. For Canadian dealers, this means: Does the tool's outbound messaging (if any) support CASL-compliant consent tracking? Does the call recording feature include disclosure language in the greeting? If you're in Ontario, is the tool aware of OMVIC advertising guidelines if it's generating any ad copy or chat responses? If you're in B.C., does it account for PIPA? These aren't hypothetical questions. Your zone rep and your OEM co-op administrator will ask them eventually.

Month 2, if you move forward: Pick one metric to move. Not five. One. If your appointment set rate from calls is 9%, make it 12% in 90 days. Run the tool, review the reports weekly, run the training, and measure that one number. If it moves, the tool is earning its keep. If it doesn't, the problem is somewhere else.

The Compliance Layer Canadian Dealers Can't Skip

This deserves its own section because it keeps coming up.

Most of these AI call analysis and follow-up tools were built for the U.S. market. The compliance landscape is different here. A few specifics:

CASL: Any AI-powered follow-up tool that sends commercial electronic messages (email, SMS) to customers needs to be operating within a CASL-compliant framework. That means consent documentation, unsubscribe mechanisms, and sender identification. If the vendor can't show you how their tool handles CASL compliance, that's a red flag.

OMVIC (Ontario): If the AI tool is generating any chat responses, scripts, or suggested language that includes pricing or financing claims, those need to comply with OMVIC's all-in pricing requirements. Under OMVIC's advertising guidelines, the advertised price must include all fees and charges the dealer intends to collect (except HST and licensing, which must be clearly excluded). An AI chat bot that quotes a price without those disclosures is creating an OMVIC compliance problem.

MVSABC (B.C.): Similar issues around price claims and certified pre-owned designations. If the AI is generating chat responses about CPO vehicles, those responses need to use the language MVSABC requires.

Competition Bureau Canada: Active enforcement on deceptive marketing practices. If an AI tool is generating claims about your vehicles or pricing that can't be substantiated, the fines are real. Up to $10M per incident for false advertising under the Competition Act.

For a deeper look at reputation management compliance, including what happened to dealers who used fake review vendors and lost their Google Business Profile for 60-90 days, see our dealership reputation management guide.

How This Connects to the Rest of Your AI Stack

AI call analysis doesn't live in isolation. It's one piece of a broader set of AI tools that are changing how dealerships operate.

Your DMS probably already has some AI features built in. CDK Drive, Reynolds & Reynolds, and PBS all have varying levels of AI-assisted reporting and customer communication tools. Before you subscribe to a standalone call analysis tool, check what your DMS already does. You might be paying for overlap. For a detailed comparison of what each DMS platform actually offers, see our DMS AI features comparison.

The OEM chat widget question is also relevant here. Most OEMs push a branded chat widget that overrides your own chat UX. If Conversica or another AI follow-up tool is also trying to manage chat, there can be conflicts. We cover that specific problem in our OEM AI chat widget vs your own guide.

And if you're thinking about AI tools more broadly for your inventory, the VDP copy generation piece is worth understanding separately. See our AI VDP copy at scale guide for how that works.

Decision Framework: Which Tool Fits Your Store

Don't pick based on the best demo. Pick based on your actual constraint.

If your problem is lead follow-up volume (BDC is dropping leads after the first touch, internet leads going cold after 48 hours): Conversica is worth evaluating. The AI follow-up automation addresses that specific problem. Just get your CASL compliance reviewed before you turn it on.

If your problem is call quality (you suspect your BDC is having bad conversations but you can't prove it): Foureyes is the more direct tool. It's built to score calls and surface missed opportunities. More affordable for a single rooftop, and the integration with your website platform (if it works) gives you the VDP-to-call connection that makes the data useful.

If your problem is marketplace lead cost (you're spending heavily on Cars.com and trying to convert more of those leads): Cars.com Conversations might help at the margin. But the honest answer is that the tool doesn't fix the underlying economics of marketplace leads. If you want to shift to first-party inbound lead flow, the answer is Google Ads and SEO, not a better chat tool on a third-party listing.

If you're a multi-rooftop group (3+ stores): None of these tools will give you clean cross-rooftop reporting out of the box. You'll need to evaluate whether the tool's reporting architecture supports consolidated views across stores, or whether you're going to be managing five separate dashboards. That's a question to ask in the demo, not after you've signed.

If you're not sure which problem you actually have: Go back to the worked math example above. Pull your baseline numbers first. The data will tell you where the leak is. Then pick the tool that addresses that specific leak.