The Speed Audit • Specialty 5 of 13
On most dealer sites, somewhere between 10 and 30 cents of every Google Ads dollar disappears into a single, measurable thing.
It is called Core Web Vitals. It is Google's measurement of how fast your website loads on a phone. The industry-wide failure rate is north of 90 percent. Yours might be in that group. It might not. The only way to know is to measure.
What this could be costing you right now
If your LCP is over 4 seconds, you are likely throwing away a meaningful share of every ad dollar. If it is under 2.5, you are not. Until you measure, you do not know which side you are on.
Largest Contentful Paint, or LCP, is how long it takes the main content of your page to load. Google measures it. Google Ads quality scores incorporate it. When LCP is bad, your bid prices inflate quietly, your Quality Score drops, your CPC creeps up, and your conversion rate falls because shoppers leave before the page paints.
Shift Digital quantified the impact for the auto industry in 2025. Dealers failing Core Web Vitals waste, on average, around 30 percent of paid search budget. Dealers passing waste close to zero. The range is wide because the inputs vary. The slider below is a directional estimate based on your spend and your current status. Not a promise about your account.
Estimate the range, then run the scan to find your real number.
Estimated monthly waste, directional
$4,500
Estimated monthly ad waste at ~30% efficiency loss on slow pages. Annualized: $54,000. Your real number depends on your specific landing pages, geos, and Quality Scores.
What nobody has told you
"Your web vendor probably has not raised this because their hosting margin moves the wrong way when they upgrade your CDN. The cheap edge tier is often the thing pushing your CWV into the red. Bringing it up costs them money."
"Your agency probably has not raised this because they sell you Google Ads, not site fixes. When CPC drifts up, the deck shows new keyword strategies. It rarely shows that Quality Score is being throttled by a 6.4-second LCP."
"Your OEM marketing co-op probably has not raised this because most co-op compliance reviews are still desktop-first and do not include CrUX field data."
"You may be in great shape. You may be bleeding. The only honest answer is to scan and find out."
The five failures we find on almost every dealer site
On most dealer sites in this area, you almost certainly have at least three of these.
Largest Contentful Paint over 4 seconds
The big image or headline at the top of your VDP takes too long to render. Google measures this on mobile, on a simulated 4G connection. The industry-wide failure rate on dealer sites is high. Whether yours is in the failing group depends on your platform, theme, and tag stack.
How we fix it: Image format conversion to WebP/AVIF, hero image preloading, CDN tier upgrade, reduce render-blocking JavaScript above the fold.
Cumulative Layout Shift over 0.1
Your page jumps as it loads. Buttons move. Images push content down. Shoppers tap the wrong thing and leave. Google penalizes the experience. Some dealer sites have CLS values in the 0.3–0.5 range; others are clean. The variance is real.
How we fix it: Reserve image and ad slot dimensions in CSS, eliminate dynamically injected content above existing content, fix font-swap layout shifts.
Interaction to Next Paint over 200ms
When a shopper taps your menu, your filter, or your CTA, the page may be slow to respond. INP measures this. Heavy chat widgets, third-party scripts, and unoptimized event handlers can push INP well past 500ms, but only on some pages, on some devices.
How we fix it: Defer or remove non-critical JavaScript, audit and replace heavy third-party tags, optimize event handlers, code-split the bundle.
Third-party tag bloat
We have audited dealer sites running 30+ tags through GTM. Chat widgets, trade tools, video chat, ad pixels, duplicated analytics. Each tag adds load time. Many are firing on every page when they only need to fire on a few. We have also seen lean stacks under 10. Yours could be either.
How we fix it: GTM audit and consolidation, lazy-load non-critical tags, remove duplicate analytics, condition tag firing by page type.
Render-blocking CSS and JS
Your page may not start rendering until the browser downloads, parses, and executes a long list of CSS and JS files in the head. Most dealer platforms ship with a long default list. Some has been pruned. Most has not.
How we fix it: Critical CSS extraction, defer non-critical CSS, async/defer JavaScript, eliminate unused CSS via PurgeCSS, inline above-the-fold CSS.
Inside the full Speed Audit
Twenty-three things we measure. Six of them you cannot see in PageSpeed Insights.
Core Web Vitals on mobile
- • Largest Contentful Paint (LCP)
- • Interaction to Next Paint (INP)
- • Cumulative Layout Shift (CLS)
- • First Contentful Paint (FCP)
- • Time to First Byte (TTFB)
Core Web Vitals on desktop
- • All of the above, separately measured for desktop
Beyond Lighthouse
- • Third-party tag inventory and individual load impact
- • Image format and optimization audit (WebP/AVIF readiness)
- • Font loading strategy (FOIT/FOUT, swap, preload)
- • Critical CSS coverage
- • Render-blocking resource map
- • JavaScript bundle analysis (size, parse time, unused code)
- • CDN coverage and edge configuration
- • Mobile vs desktop divergence
Real-user data (where available)
- • Field data from CrUX (Chrome User Experience Report)
- • Mobile vs desktop split in the field
- • Geographic variance
SAMPLE AUDIT OUTPUT
Poor. Easy wins available
- Third-party tags 31
- Render-blocking JS 12 files
- Image format JPEG-only
Estimated monthly waste
$7,500
Top fix: Hero image preloading. Estimated lift: +18 score points.
Illustrative sample. Not your site.
What happens when you fix it
Fixing this is not a website redesign. It is a series of specific, surgical changes.
01
DIAGNOSIS
We give you the report
48 hours. Every finding, ranked by impact and effort, with platform-specific instructions for your CMS. You can hand it to your web vendor and they can implement most of it.
02
IMPLEMENTATION (OPTIONAL)
We do the work
If your vendor cannot or will not, we install the fixes ourselves. Most live in Google Tag Manager, the head of the page, or as a CDN-level configuration. Live in 7 to 14 days.
03
VERIFICATION
We re-run the scan
30 days after the fixes go live, we re-run the audit and confirm the lift. Site Health Score increase, category-level pass, downstream impact verified.
What the data says.
~30%
estimated share of dealer paid-search spend wasted on pages that fail Core Web Vitals. Industry average. Your share could be much higher or close to zero.
Source: Shift Digital, 2025.
>90%
of dealer websites we have measured fail Core Web Vitals on mobile at least once a quarter. Whether you are in that group on the day Google looked depends on traffic patterns and recent changes.
2.5s
is Google's threshold for a 'Good' mobile LCP. The industry average we see on dealer VDPs is meaningfully higher. The variance between platforms and themes is large.
The Proof
How five speed fixes lifted Core Web Vitals and dropped cost per lead on a franchise dealer site.
We deployed five speed fixes on a franchise GMC dealer site running on Dealer.com. Going in, their LCP on mobile was 6.7 seconds. Their CLS was 0.34. They had 31 third-party tags firing through GTM, including 4 different analytics scripts that duplicated each other.
We deployed five speed fixes: image format migration to WebP, hero image preload, GTM tag consolidation (cut from 31 to 14), font loading strategy (preload + display swap), and inline critical CSS. Total developer hours: 12. Total deployment time: 6 days.
Mobile LCP dropped from 6.7s to 2.1s. CLS dropped to 0.04. Their PageSpeed score on mobile went from 38 to 84. Their Google Ads Quality Score lifted across their top 50 keywords. Their cost per lead from Search dropped 22 percent over the next 60 days. Same campaigns. Same budgets. Just a faster site.
Speed was not the only lever we pulled with this dealer. But it was the first.
6.7s → 2.1s
Mobile LCP, before and after
38 → 84
PageSpeed Score, mobile
31 → 14
Third-party tags firing
-22%
Cost per Search lead
12 hours
Total developer time to deploy
Pricing
Three options. No retainer. No negotiation.
Pay once for the diagnosis. Pay once for the fixes. Or pay once and get both.
$997
Full Diagnostic
- •Comprehensive site audit across all 14 specialty areas
- •Search Console authenticated review
- •Platform-specific fix instructions
- •Priority-ranked master fix list
- •Estimated impact and effort per fix
- •Refundable if we do not find 3 surprises
- •48-hour delivery
$6,497
Fix Sprint
- •Bring your own audit report
- •We implement your top 5 priority fixes
- •Schema, speed, and structural fixes
- •Live within 30 days
- •No code changes required from your vendor
$6,997
Full Treatment
- •Full Diagnostic included ($997 value)
- •We implement every fix on the priority list, not just 5
- •Schema, speed, and structural fixes
- •Live within 7 to 14 days
- •90 days of regression monitoring after launch
- •We handle your web vendor directly
$500 more than the Fix Sprint. Includes the $997 diagnostic and the complete fix list.
Speed Audit-specific questions dealers ask.
Answered straight.
How is the free Speed Scan different from PageSpeed Insights?
PageSpeed Insights is a free Google tool that runs Lighthouse against any URL. It gives you scores. We use the same underlying APIs but we add three things PageSpeed cannot give you: dollar-cost attribution to your specific ad spend, fix instructions specific to your dealer platform, and the third-party tag inventory PageSpeed does not produce.
What is a passing Core Web Vitals score?
Google's thresholds are: LCP under 2.5 seconds, INP under 200 milliseconds, CLS under 0.1. To pass, you need all three at or below those thresholds for at least 75 percent of your real users on mobile.
Does speed actually affect my Google Ads cost?
Yes, directionally. Quality Score considers landing page experience, which includes Core Web Vitals. Lower Quality Score means higher CPC and lower ad rank for the same bid. The magnitude varies by account. The directional effect is well-documented.
Why might my site be slow if I am on a major dealer platform?
Major dealer platforms ship with shared frameworks, default themes, and aggressive third-party tag stacks. The platform itself is rarely the bottleneck. The configuration usually is. Most dealer accounts have accumulated tags and modules that nobody removed.
Do I need to redesign my site to fix speed?
Almost never. In our experience, most speed gains come from changes that do not require a redesign: image format migration, GTM tag consolidation, font loading optimization, critical CSS, and CDN tier upgrades.
How long does the free Speed Scan take?
Thirty seconds in the browser. The full report with fix instructions is emailed within 60 minutes.
How accurate is the dollar-cost estimate?
It is an estimate based on industry research applied to the spend you tell us about. It is designed to be defensible, not flattering. The actual savings on your specific account will vary.
Will fixing speed affect my conversion rate?
Almost always positively, with magnitude depending on starting point. Sites with mobile LCP over 5 seconds typically see meaningful conversion lift after fixes. Sites already in the 2–3 second range see less.
What about my chat widget? Will I have to remove it?
Probably not. Most chat widgets can be lazy-loaded so they only load after the user is ready to interact.
What if my web vendor refuses to implement the fixes?
Most of our speed fixes can be implemented at the GTM layer, the CDN layer, or via tag injection that does not require vendor cooperation. For the rest, we provide a vendor-ready spec sheet, or we implement on the Implementation tier ourselves.
Do you work with my platform?
If you are on Dealer.com, DealerOn, Dealer Inspire, CDK Global, AutoTrader, or Dealer eProcess, yes. The fix instructions are platform-specific.
How long do speed fixes typically last?
If nobody touches the site, indefinitely. In practice, dealer sites accumulate new tags and modules every quarter. We recommend a re-scan every 6–12 months.
While we are at it
Speed Audit is rarely the only thing wrong. Three areas dealers usually pair with this one.
The Tracking Audit
GA4, GTM, Enhanced Conversions, Offline Conversion Import. If tracking is broken, every other decision is guesswork.
Read the detailsThe Schema Audit
AutoDealer, Vehicle, LocalBusiness, FAQ. Missing or malformed schema costs you AI search and rich-result eligibility.
Read the detailsThe Domain Audit
HTTPS, redirects, DNS hygiene, SSL chain, domain age signals. The boring stuff that silently throttles everything else.
Read the detailsFind out where your speed actually stands.
30-second scan. We score your Core Web Vitals, surface the three biggest leaks, and give you an honest read on whether this is a real problem on your site or not.