Why Every Vendor Promises ROI (And How to Verify It)

Most dealers rank ROI as their top criterion when evaluating new technology. Vendors know this. That’s why every pitch deck you’ve sat through in the last three years opens with a slide showing revenue going up and to the right. The number changes. The story doesn’t.

Every dealership tech vendor claims ROI, but fewer than 1 in 5 can show net returns after subtracting all costs, including labor and baseline performance. The five most common tricks are cherry-picked time periods, uncontrolled before-and-after comparisons, gross-only math, attribution gaming, and projections presented as actuals. Use the verification checklist below before signing anything. — Based on NADA and Cox Automotive dealer survey data

It sounds like you’ve been burned before. Maybe not badly enough to cancel, but enough to stop trusting the next pitch. You sat through the demo, the numbers looked great, and six months in, your controller couldn’t find the lift anywhere in the financials. You’re not alone. Cox Automotive’s 2024 dealer sentiment study found that 61% of dealers feel overwhelmed by the number of technology vendors in the market. That’s not confusion. That’s exhaustion from being sold to.

This article is an accusation audit. We’re going to say the quiet part out loud: vendors, including us, have an incentive to make their numbers look as good as possible. Then we’re going to give you a framework for separating real data from vendor math.

The 5 ROI Tricks Vendors Use on Dealers

These aren’t necessarily lies. Some vendors genuinely believe their numbers. But belief doesn’t make data valid, and good intentions don’t fix bad methodology. The average dealership spends hundreds of thousands of dollars per year on technology and services. That’s too much money to evaluate on trust alone.

Want AI that does something useful for managers? Try the live demo and see how Ringlead connects leads, scores calls, and flags deals that need attention.

1. Cherry-Picked Time Periods

A vendor shows you a dealer who gained 30 units in their first 90 days on the platform. What they don’t show you is the six months after that, when the number settled back to a 10-unit gain. Or the seasonal context: those 90 days happened to be March through May, when every store in the country sells more cars.

Cherry-picking isn’t always intentional. Sales teams genuinely remember their best stories. But the effect is the same. You’re making a 12-month commitment based on a 90-day window.

What to ask: “Can I see 12-month rolling data for at least five stores in my market size?“

2. Before-and-After Without Controls

This is the most common trick in automotive tech. “Store X was at 150 units. They started using our platform. Now they’re at 190 units.” The implication is that the platform caused the lift. But during that same period, maybe the store hired a new GSM who rebuilt the sales process. Maybe a competing store two miles away closed. Maybe they increased their ad spend by $40,000 per month.

A valid before-and-after comparison isolates the variable. In academic research, this is called a controlled study. In vendor marketing, it’s called inconvenient, which is why you almost never see it.

What to ask: “What else changed at that store during the ‘after’ period?“

3. Gross Numbers Instead of Net

A vendor tells you their platform generated $200,000 in additional gross profit last year. Sounds great. But the platform costs $36,000 per year. Your BDC coordinator spends 8 hours per week managing it, which is another $12,000 in loaded labor. Training your team took two Saturdays. And 40% of those “additional” deals would have happened anyway because the customers were already in your CRM pipeline.

When you subtract all of that, the $200,000 becomes $60,000. Still positive, but a very different story than the one in the pitch. Foureyes research on mishandled leads shows that 43% of leads get mishandled at the average store, which means there’s real money on the table. But the size of the pile and the size of what a specific tool recovers are two different numbers.

What to ask: “What’s the net ROI after subtracting platform cost, labor to manage, training time, and baseline performance?“

4. Attribution Games

Attribution is where vendor math gets truly creative. A customer submits a lead on Monday. Your speed-to-lead tool connects them to a salesperson in 40 seconds. The customer says they’re not ready. Two weeks later, they walk in from a Google ad, test drive, and buy. The speed-to-lead vendor claims the deal. The Google Ads vendor claims the deal. Your CRM vendor claims the deal because it was “in the system.”

That one deal just generated three ROI claims from three different platforms. According to Google’s own automotive research, the average car buyer has over 900 touchpoints before purchase. Every vendor touching any of those 900 moments can technically claim attribution.

What to ask: “How do you handle attribution when multiple tools touch the same customer? Do you use first-touch, last-touch, or multi-touch?“

5. Projections Presented as Actuals

This is the subtlest trick. A vendor shows you a slide: “Based on your current lead volume and our average lift, you should see an additional $15,000 per month in gross profit.” The word “should” is doing all the heavy lifting. That’s a projection, not a measurement. It’s built on averages from their best-performing stores, applied to your specific situation, with no guarantee it’ll translate.

Projections aren’t worthless. They’re starting points. But when the projection shows up in the contract negotiation as if it’s a promise, something has gone wrong. CDK Global’s research suggests that only about 68% of dealers who adopt new technology report positive impact, which means roughly a third don’t see the results the projection predicted.

What to ask: “Can you show me actuals from stores similar to mine, not projections based on averages?”

The Vendor ROI Verification Checklist

Before you sign your next annual contract, run the vendor’s claims through these seven questions. You don’t need a finance degree. You need 20 minutes and a willingness to make the salesperson uncomfortable.

#	Question	Red Flag If…
1	Show me net ROI after all costs including labor	They only quote gross numbers
2	Show me 12-month rolling data, not a single quarter	They offer “representative examples” instead
3	What’s your renewal rate after 12 months?	They won’t answer or say “we don’t track that”
4	What’s your cancellation rate?	They pivot to “customer satisfaction scores”
5	Can I pick my own references from your client list?	They insist on providing pre-selected references
6	What else changed at the stores in your case studies?	They can’t name a single confounding variable
7	Are these projections or actuals?	They blur the line between the two

If a vendor passes all seven, they’ve earned a second meeting. If they fail three or more, the data in their pitch isn’t reliable enough to base a purchasing decision on. That doesn’t mean the product is bad. It means the evidence isn’t there yet.

Why Ringlead Is Telling You This

It sounds like you’re wondering why a vendor would write an article teaching you to question vendors. Fair question. We’re writing this because we’re tired of cleaning up the mess that bad vendor math creates in the market.

When a dealer gets burned by a tool that promised 30 extra units and delivered 4, they don’t just cancel that vendor. They stop trusting the next one. And the next one might actually work. The state of dealership AI adoption shows that fewer than 15% of dealers have moved beyond basic chatbots, partly because the vendors who came before us over-promised and under-delivered. One area where you can verify results without trusting a vendor at all: AI search optimization, where you can see directly whether ChatGPT and Perplexity cite your dealership when shoppers ask for recommendations.

Ringlead is a vendor. Everything in this article applies to us. We’d be hypocrites to pretend otherwise. The difference is what we measure. Ringlead tracks speed-to-lead on your actual leads and uses AI to score your actual calls. The data is yours. We don’t project what could happen from a model. We measure what is happening right now and show you the gap. If your response time is already under 60 seconds and your call quality is strong, we’ll tell you that you don’t need us.

That’s not altruism. It’s math. Dealers who stay because the data is real stay for years. Dealers who sign based on projections cancel in six months. We’d rather have the first kind.

How to Evaluate Any Vendor Demo Going Forward

The verification checklist works after the pitch. But you can save yourself time during the pitch by watching for three signals.

Signal 1: They won’t show you the bad. Every platform has stores that didn’t work out. If a vendor claims 100% success, they’re either lying or they haven’t been around long enough to have failures. Ask: “Tell me about a store that cancelled and why.” A vendor who answers honestly is a vendor who knows their product’s limits.

Signal 2: They compare themselves to doing nothing. “Without our tool, you’re losing $X.” That’s not a comparison to the alternative. That’s a comparison to zero. The real question is whether their tool outperforms the next best option, which might be a cheaper tool, an internal process change, or just hiring one more person. For a framework on comparing categories, see BDC services vs technology.

Signal 3: The ROI calculator has no inputs for your costs. If a vendor’s ROI calculator asks for your lead volume and average gross but doesn’t ask for your labor costs, training time, or current baseline performance, it’s not calculating ROI. It’s calculating gross revenue attribution, which is a very different number.

Try the Live Demo

Frequently Asked Questions