#What is actually loading
Open a dealership vehicle detail page and look at what runs before the page becomes usable. There is typically a chat product, sometimes two. A trade valuation widget. A credit application tool. An inventory merchandising script. Call tracking, which rewrites phone numbers on the page after it loads. Session recording. Retargeting pixels for however many agencies have been engaged. The platform's own analytics. And a tag manager containing several more that nobody has opened in a year.
Each one was added by somebody with a good reason. None was ever removed, because removal requires knowing what it does and who asked for it, and that information left with the person who signed it.
Images, by contrast, are the part everyone blames and the part every modern dealer platform already handles. They are lazy-loaded, they arrive after the page is usable, and on a well-built template they are not the thing keeping a shopper waiting.
#Why nobody removes any of it
The structural reason is that no single party is accountable for the total. Ask any individual vendor whether their script is the problem and the honest answer is no — theirs is small. All of them are right, and the page is still slow, because the total is the sum of decisions that were never made together.
The second reason is fear. A tag nobody understands might be the one feeding a report somebody looks at, or the one a manufacturer requires. Removing it feels riskier than leaving it, so it stays, and next quarter another one joins it.
The third is that page speed has no owner. It is not the website provider's problem because they did not add most of the scripts. It is not the agency's problem because they added one. It is not the general manager's problem because it is technical. So it belongs to nobody, which is the same failure pattern that produces every other gap described on the car dealer marketing hub.
#The audit, which costs nothing but a morning
List every script firing on a vehicle detail page and on the homepage. Your browser's developer tools will show you, and so will your tag manager, and the two lists will not match — the difference is usually scripts injected by the platform or by another script, which is itself worth knowing.
For each one, answer two questions: which vendor is this, and are we still paying them. Anything that fails the second question comes off immediately and nobody will notice. Stores routinely find pixels from agencies they parted ways with two or three years ago, still loading on every page view.
For what remains, ask which report anybody actually reads. Two chat products is common and one of them can go. Two call tracking systems is common. Session recording that nobody has watched in six months is a page-speed cost and, increasingly, a privacy exposure — which is a separate problem covered in dealership website privacy demand letters.
TABLE
The script inventory: what is on a dealer page and what to do with each
Work down this list against your own vehicle detail page. The third column is where the free wins are — every one of them is a removal nobody will notice.
| What is running | Why it was added | How to decide | Commonly duplicated |
|---|---|---|---|
| Chat | Somebody wanted after-hours coverage | Which one do the salespeople actually watch? Keep that one. | Yes — a platform chat and a vendor chat, both live |
| Trade valuation widget | A lead source with its own dashboard | Count the appraisals it produced last quarter, not the submissions | Sometimes |
| Credit application / digital retailing | A vendor pitch that made sense at the time | Applications started and completed, not clicks on the button | Rarely, but often unused |
| Call tracking | Attribution for paid media | One system, owned by one party. Two systems rewrite each other. | Yes — frequently one left by a previous agency |
| Session recording | Somebody wanted to see where people drop off | Has anybody watched a session in six months? | Rarely, but often abandoned |
| Retargeting pixels | One per agency, ever | Match each to an agency you currently pay. Remove the rest today. | Yes — this is where dead vendors live |
| Inventory merchandising | Part of the platform or an add-on | Usually not removable; ask which parts are optional | No |
| Platform analytics | Included, not optional | Sets your speed floor. Ask what it costs you in load time. | No |
| Tag manager container | The place everything above got added | Open it. The contents are usually a surprise. | The container itself is the duplication risk |
Categories compiled 2026-09-02 from the script layers recurring on dealership vehicle detail pages. What is on your own site is a measurement, not an assumption — take the list from your browser's developer tools.
#What speed actually costs you
The visible cost is people leaving before the page renders, and it is worst on the traffic you paid for — a shopper arriving from an ad on a phone on a weak connection is the least patient visitor you have, and the most expensive one to lose.
The second cost is search visibility, and it is more subtle than the ranking-factor framing suggests. A slow site is not merely penalised; it is crawled less efficiently, which matters enormously for a site with tens of thousands of inventory URLs and a limited crawl budget. Time spent fetching a slow page is time not spent finding your service page.
The third cost is that it makes everything else you buy work less well. Every channel lands here. A store spending heavily on advertising into a site that takes eight seconds on a phone is paying for the click and then discarding it, which is the least examined line in most dealership budgets.
#What the platform can and cannot fix
Some of the load is the platform's own and cannot be removed — its analytics, its merchandising layer, sometimes its chat. That is a real constraint and it sets a floor. Ask which of the platform's scripts are removable and get the answer in writing; it is one of the eight questions in the dealership website selection table for exactly this reason.
Everything above that floor is yours, and it is normally the larger share. That is the good news in an otherwise unglamorous problem: most of the fix does not require the platform's cooperation, a redesign or a budget.
Where the platform genuinely is the constraint — inventory rendered entirely in the browser, a template that cannot be trimmed — that is a migration argument rather than an optimisation one, and it should be made on that basis rather than on the homepage looking dated.