Spiffy Speed Script · one head script
Slow site? Fixed.
Nothing rebuilt.
Faster load times = lower bounce rates & more lead submissions
Built for dealers who like their website but need it faster.
Measured delta
captured 2026-07-22Dealer group flagship site - South · Lighthouse mobile · script off vs on +32 points
One script tag, end to end
The whole intervention, drawn to scale: your site stays yours, one tag goes in the head, the edge does the work, and the verdict comes from field data, not our lab.
01 TAG One line in the head of the site you already have - no replatform, no redesign.
02 EDGE The edge rewrite streams your HTML and defers what blocks rendering.
03 VERDICT Judged on 28-day CrUX field data, not lab screenshots - wins and losses published.
What one tag did
Measured on live client sites, script off vs on - every one of the four moved.
Google Lighthouse mobile, script off vs on, multi-run average per site
Honda store - Southern California · LH mobile
Multi-brand dealer group - Midwest · LH mobile
Multi-brand dealer group - South · LH mobile
Dealer group flagship site - South · LH mobile
4 client dealerships - results published; identities shared privately on request.
Per-store receipts - run ranges and field dataTwo states. One variable.
We measure the same page with the script off and on, on Google's own tools, and publish the result either way.
Dealer group flagship site - South · LH mobile · multi-run average
32
64 +32
same page, script off vs on - the only variable is the tag · bar length = LH score
The win - the head script on live client sites
4 client dealerships, Lighthouse +15 to +32 points each · Google Lighthouse mobile, script off vs on, multi-run average per site
Lab · what screenshots show
A synthetic run in a clean lab - it can be inflated by moving work past the measurement window, not by removing it.
Field · what Google ranks on
28-day real-user Chrome data (CrUX) - no script can game it, and it's the number that moves rankings.
The dud - a competitor's bolt-on script, isolated
12 Lighthouse runs · median-of-3 · allowed vs blocked
It relocates main-thread work; it doesn't remove it.
The wash - our rewrite on a JS-built page
client-rendered homepage · 161 scripts · 53 stylesheets
When JavaScript assembles the page after load, static optimization can't finish the job, and we say so.
win = client ledger (script off vs on, multi-run average per site), captured 2026-07-22 - client results on /spiffy/clients · dud = competitor script isolated allowed-vs-blocked · wash = our own edge-rewrite fair A/B - full method on /spiffy/proof
Where Spiffy fits
Speed Script · you are here
One script in the head of the stack you already run, and we tell you up front where a bolt-on can move the needle and where only architecture can.
Dealer Websites
A ground-up dealership site compiled to pure HTML/CSS/JS, served from the Cloudflare edge - the ceiling, when you're ready for it.
Explore Dealer WebsitesWe measure first. You decide after.
Request a demo and we'll measure the site you have today - script off vs on - and show you the delta before you commit.