01

What does the speed analysis cover?

  • Homepage, search-results, vehicle-detail and conversion pages
  • Mobile and desktop laboratory testing
  • Available real-user Core Web Vitals field data
  • Third-party scripts, network requests and loading sequence
  • Visible experience, interaction delay and layout stability
02

What does the analysis not do?

The analysis does not assign blame from one screenshot, promise a ranking increase or recommend removing valuable dealership tools simply to create a cleaner test.

Its purpose is to identify the real loading constraints and determine whether Speed Engine is a useful fit.

03

What is the next step?

Submit the dealership URL. The initial analysis will establish the representative page set and the performance conditions that should be measured. If you would rather see a number before talking to anyone, the free speed test runs on the same public data and asks for no email address.

04

What does a thorough speed analysis look at?

  • Field Core Web Vitals for the domain, mobile and desktop reported separately
  • Representative pages: homepage, new and used search results, and multiple vehicle detail pages
  • Every third-party provider on those pages, ranked by main-thread blocking time
  • Image weight and whether the largest element is being delayed by lazy loading
  • Layout stability, and which specific elements are moving after render
  • Render-blocking resources in the critical path
  • Duplicate and orphaned tags left behind by ended campaigns
  • Whether tools depend on each other in ways that constrain sequencing
05

What will the analysis not tell me?

The audit will not give you a predicted percentage improvement. That number cannot be known before a change is made and measured on your own traffic. Any vendor supplying one is estimating and presenting it as a finding.

It will not blame your platform vendor by default. Sometimes the platform is a genuine constraint; more often it is not, and saying so would be easier than being accurate.

It will not treat a single laboratory run as evidence of what your shoppers experienced.

06

How should a dealership use the findings?

  1. 01

    Take the dead weight first

    Tags nobody owns and vendors no longer under contract. No trade-off, no negotiation.

  2. 02

    Take the findings to your vendors

    A provider consuming several hundred milliseconds of main-thread time on every page view is a conversation worth having, and most vendors have options they do not volunteer.

  3. 03

    Decide what genuinely must load early

    Consent management and anything a shopper touches immediately. Write the list down so it survives staff changes.

  4. 04

    Re-measure after four weeks

    Field data moves on a 28-day window. Judging sooner means judging noise.

07

What is checked on each page type?

Coverage by dealership page type
Page typeWhy it mattersWhat is checked
HomepageBrand entry point and often the only page testedField vitals, hero image, render-blocking resources
Search resultsHigh traffic, repeated images and filtersImage handling, filter responsiveness, third-party load
Vehicle detailClosest page to a lead, heaviest tool stackGallery handling, layout stability, form path, all third parties
Conversion pagesWhere a lead is submittedForm function, interaction delay, anything the lead path depends on
08

What does the analysis look at?

The starting point is the page set a shopper actually uses. That means the homepage, at least one populated inventory search-results page, and several vehicle detail pages chosen to span the range of media weight on the site. It also means the conversion paths: service scheduling, the finance application, or whichever lead route the dealership cares most about.

For each of those the analysis records what the field data says, where it exists, and what the laboratory trace says about why. The two are reported separately, because a laboratory score presented as a real-user result is the most common way this measurement is misused. Then the third-party inventory: which providers are present on which templates, how long each occupies the main thread, and which of them the page genuinely needs before it becomes useful.

The output is a description of the constraints on that specific website, and a straight answer about whether Speed Engine is a useful fit for them. Sometimes it is not. A site whose problem is a 4 MB uncompressed hero image has a cheaper fix available, and saying so is more useful than selling around it.

09

What happens after the request?

You get a reply from a person who has looked at the site, not an automated report. The first exchange establishes the representative page set and the conditions to measure under, because a review of the wrong pages is worse than no review.

There is no obligation attached, no platform change proposed, and no requirement to remove any tool. If the honest answer is that the dealership's constraint is something this product does not address, that is the answer you will get.

Five ordered stages of a dealership website speed analysis: the request arrives, a page set is chosen that includes inventory and lead-driving pages, field and laboratory evidence are read separately rather than averaged, the resources competing during load are inventoried, and every finding is returned with an owner named. The drawing shows order only and reports no measurement.
The order the analysis runs in. Sequence only: nothing here reports a measurement or a finding for any specific website.

Sources and further reading

External sources support the general technical guidance on this page. They do not represent a DealerSpeed Engine performance result.