Most dealer Google Ads accounts have a leak problem. Not a strategy problem, not a creative problem, not even an account-structure problem in most cases. Just leaks — places where money quietly drains out of the account every day, invisible to the weekly budget review because they don’t show up in any of the standard reports the GM looks at.
Audit a hundred dealer accounts and the same five leaks appear in roughly eighty of them. Together they typically waste 20-30% of monthly spend, which on a $25,000-a-month account is enough to fund an entire fifth-channel rollout if you stopped the bleeding. Most dealers don’t, because the leaks aren’t catastrophic enough to trigger a panic — they just degrade performance in a way that looks like the market getting harder rather than the account being broken.
Here are the five leaks, in roughly the order they’re worth fixing, and the practical way to plug each one.
Leak #1: Negative keyword neglect
The most common, the most expensive, and the most boring to fix. Dealer accounts almost universally under-invest in negative keywords.
Three negative-keyword categories show up in nearly every audit. First, service-related searches. If your dealership runs a service department, you don’t want your sales campaigns paying for clicks on “oil change near me,” “transmission repair,” or branded service searches that the service department isn’t tracking back. Service traffic should be its own campaign with its own budget, and the sales campaigns should aggressively negative every service-intent term.
Second, OEM head terms with the wrong intent. A Ford dealer running a campaign on “Ford F-150” as a head term is paying premium CPCs for a search that’s probably someone configuring a build on the Ford site, not someone shopping locally. Negative out the broadest OEM terms and let your model-specific, local-intent campaigns catch the actual buyers.
Third, jobs and careers. Dealer brand searches frequently include “jobs,” “careers,” “hiring,” and similar — and unless you’ve built that specifically as a recruiting funnel, those clicks are pure waste.
Fix: pull the search-terms report for the last 90 days, sort by spend, and aggressively negative every term that drove cost without conversion. Then build a shared negative-keyword list across all sales campaigns so you don’t fix it once and let it drift back.
Leak #2: Match-type drift
Google has spent years quietly pushing every account toward broad match. Each year the platform’s recommendations and account scripts nudge advertisers off exact and phrase match toward broader matching, on the promise that machine learning will find the right queries.
For most dealer accounts, this is a slow leak. Broad match without aggressive negative-keyword discipline lets the account match against tangentially related searches that look like they should convert and don’t. You end up paying for “best truck under $50K” when you sell only used SUVs, or “Carfax report” when you’re a dealership.
The fix isn’t to refuse broad match — it has its place — but to enforce a structural rule: any campaign running broad match must have a robust negative-keyword list and a tight conversion target. If both aren’t true, the campaign should be on phrase or exact match. Most dealer accounts have broad-match campaigns running with the negative-keyword discipline of an exact-match campaign, and the cost shows up in the search-terms report.
Leak #3: Vehicle Listing Ads placement and feed quality
Vehicle Listing Ads have become a meaningful share of dealer spend, but the optimization gap between a well-tuned VLA setup and a default one is dramatic.
The two biggest leaks here are feed quality and placement coverage. Feed quality issues — wrong fitment data, missing trim levels, image-resolution failures, price discrepancies between feed and VDP — cause inventory to either not surface at all or surface at lower auction priority. Most accounts have somewhere between 5% and 15% of their inventory feed silently disqualified, which means the campaign is bidding only on the inventory that made it through, often the wrong slice of the lot.
Placement coverage matters because VLAs appear in multiple Google surfaces — Search, Maps, the dedicated vehicle ads on result pages — and accounts that haven’t explicitly opted in to all available placements are leaving impression share on the table. The default settings are not the optimal settings.
Fix: run a feed audit monthly. Pull the disapproved-products report, fix the obvious issues, and confirm placement coverage is fully enabled. The agencies running a Google Ads playbook tuned for dealer accounts have made this audit a standing item, not a quarterly project, because feed quality drifts continuously.
Leak #4: Geographic overlap between campaigns
Dealers with multiple campaigns — different inventory types, different OEMs, certified pre-owned versus general used, financing offers versus retail — frequently end up with overlapping geographic targets. The same buyer in the same zip code can match against three of your own campaigns simultaneously, which means you’re bidding against yourself, paying premium CPCs for clicks you would have gotten cheaper if only one of your campaigns was eligible.
This is especially common after account expansions. A dealer adds a new campaign for a specific promotion, doesn’t tighten the geo on the existing campaigns, and now both run. The new campaign looks like it’s working. The old campaign’s CPCs creep up. Nobody connects the two events because the dashboards are looking at each campaign in isolation.
Fix: audit geographic targeting across all active campaigns once a quarter. If two campaigns are eligible for the same zip codes and the same query intents, decide which one should win and exclude the others from that overlap.
Leak #5: Misattributed phone calls and false conversions
This one’s the most awkward to fix because it requires admitting some of your “conversions” weren’t conversions.
The two most common false-conversion leaks are short-duration call-tracking conversions and recycled lead-form fills. Call-tracking platforms typically count any call over 30 seconds as a conversion. The reality is that a 35-second call is almost never a sales-qualified conversation — it’s a wrong number, a service inquiry, or someone confirming hours. If your account is optimizing toward “calls” and a meaningful share of those calls are noise, the algorithm is being trained on the wrong signal.
The lead-form version is similar: a customer who’s filled out a form three times in 60 days is one customer with a frustrated salesperson, not three new leads. If your CRM is sending each one back to Google as a separate conversion, you’re inflating the conversion volume and degrading the algorithm’s targeting.
Fix: tighten the conversion definitions. Call conversions should require 60+ seconds, ideally with sales-only call routing during business hours. Lead-form conversions should deduplicate within a 30-day window. The conversion volume will drop. The conversion quality — and over time, the cost per actual sale — will improve significantly.
How to actually run this audit
The full audit takes roughly a half-day for someone who knows the account. The order matters: fix the negative keywords and conversion definitions first, because those are training the algorithm wrong every day they’re broken. Match types and geo overlap can be cleaned up second. Feed quality is ongoing.
The honest read on most dealer accounts is that they don’t need a strategy overhaul. They need a hygiene pass. The strategy was probably reasonable when it was set up. What’s happened in the eighteen months since is drift — small unforced errors compounding while the GM watches the topline number and the agency or in-house team focuses on creative and bidding.
If your cost per sale has been creeping for the last two quarters and the explanation has been “the market,” the leaks are worth checking before you accept that explanation. Most of the time, eighty percent of what feels like a market problem turns out to be a negative-keyword problem and a feed problem hiding inside a campaign that nobody’s looked at closely in too long.
DealerSmart’s team and other operators close to dealer accounts have been making this point with increasing frequency: the gap between an audited account and an unaudited one is wider than the gap between a great strategy and a mediocre one. The fix is unglamorous. The savings aren’t.
