Celtra Integration for Car Dealers

Enterprise creative production across many markets, and a frank account of what a group would need before asking.

Celtra is creative production infrastructure for organisations that make a very large number of advertisements. A brand team builds a master design system, defines which elements vary, connects a data source, and the platform renders every size, language, market and offer combination that the plan requires, with approval workflow and version control wrapped around it so a legal or manufacturer review can happen once rather than four hundred times.

The dynamic feed support is the piece automotive people usually ask about. Given a structured source of records, the platform can render variants per record, which is how a retailer with thousands of products keeps its creative current without a designer touching each one.

What it is not is a media channel. Nothing here decides who sees an advertisement or what an impression costs. The output moves into whichever networks the group buys, on the group's own accounts and invoices. For a dealership evaluating a list of advertising platforms, that boundary is the single most useful thing to hold on to, because the tooling looks like an advertising platform in every screenshot and is not one.

The case appears when the number of variants stops being countable. A group with thirty rooftops, four manufacturer brands, market specific offers and a compliance review on every price is producing a genuinely large creative matrix every month, and doing that by hand either costs a studio or quietly does not happen. Where it does not happen, stores end up running whatever creative is left over from last quarter, which is a real and unmeasured cost.

Governance is the second argument and it is underrated in franchise retail. A system where the brand rules live in the template rather than in somebody's memory means a manufacturer's requirements are enforced by construction. A group that has ever had a co-op claim rejected over a logo treatment understands what that is worth.

The honest objection is that this is enterprise software and it behaves like it. It expects a person who owns the design system, a data source that is clean and consistent, and a commercial commitment that a single rooftop will not clear. Buying it without that owner produces a very expensive template nobody maintains, which is a worse outcome than the studio invoice it replaced.

The second objection is sequencing. Creative volume is rarely the binding constraint on a store that is not selling enough cars. Offer, media weight, trade area and follow up speed usually are, and those come first. A group whose creative is stale because nobody has time is a real customer for this. A group whose leads are not being worked has a different problem entirely, and what to expect from car sales leads is the more honest starting point.

Not started, and the test that comes with that status is checkable rather than decorative: no adapter file exists in the ad gateway for this platform. No authorization URL, no credential block, no wire-up checklist, and nothing partially implemented behind a badge.

Access class is PARTNER, and that gate is the vendor's rather than ours. Credentials come through an application the vendor reviews, and an enterprise platform reviews with commercial fit in front of it: what you would spend, how many seats, how the account would be supported. That is a slower and more selective process than opening a developer account, and it sits in front of any engineering we might do.

There is a structural point that the status word cannot carry, and it belongs here. The gateway is designed around seven operations that assume a network selling impressions. A creative production platform answers two of them naturally and has no meaningful answer for the rest. Connecting it is therefore a design question as much as a build, and a request that says what a store wants to happen concretely is far more actionable than one that names the vendor and stops.

Described as a design for something that does not exist, with a shape unlike the media platforms elsewhere in this catalogue.

The first and last rows have to be read together. Creative could be produced from vehicle data already held in the CRM, which is a genuine capability. It would not be produced from a live catalog handed to an advertising network, because we emit no such feed anywhere. The difference is freshness, and on a page about dynamic creative that difference is the whole argument.

An unbuilt integration can be described this precisely because the structure it would join already runs and already carries a live network.

One library inside LeadLocate defines what any advertising platform must be able to do: hand us an authorization URL, exchange the credential for a usable token, list the accounts that credential reaches, deploy a campaign, pause or resume it, return spend and results, and prove on demand that the connection still works. Seven operations, one adapter file per network. That single contract is why a hundred entry catalogue does not require a hundred separate architectures.

A production platform fits it partially, and pretending otherwise would be the kind of tidy claim this directory exists to avoid. Authorization and verification map across cleanly. There is no campaign to deploy, no spend to read, nothing to pause, so an adapter would implement what applies and return a refusal naming the reason for the rest, exactly as every unimplemented network does today rather than returning a silent success.

What carries over unchanged is the part a franchise compliance review cares about. Credentials are encrypted with AES-256-GCM before storage. Every action against any connected service is written to an append only audit log carrying the company, the service, the user and the detail, timestamped in UTC because the table is shared across dealerships and a company local timestamp cannot be reconciled later. Objects record which credentials created them and re-resolve those credentials afterwards, so a disconnected account never strands what was built beneath it.

Two ways to advertise through LeadLocate, and the group chooses.

The first is the webXintel Ads Network: included, built and operated by our team from the selections you make in Leads Manager, and billed as one line rather than a stack of platform statements. Creative on that path is produced by our team as part of it, so a store on the managed path is not the store shopping for this platform.

The second is your own account. Connect the accounts your group already holds and work routes through them, on your billing, in your name, with your negotiated terms intact. For this platform neither path exists today, because no adapter has been written and the vendor gate has not been approached.

One limit runs across the whole gateway and is better said here: deploying, activating, pausing and reading spend is done by our team rather than from a control on your dashboard. That is deliberate while the connection layer is young, and it is on the list to open up. A creative platform sits beside that constraint rather than under it, since nothing it renders spends money by itself.

Four steps, two of them ours. A request from a group whose creative volume genuinely warrants it. A queue position earned against every other request in the catalogue. An adapter written against the operations that apply, with a credential block, a verification path and a mapping from the vehicle attributes we hold to the variable fields in a design system. And a vendor reviewed application for access, on the vendor's timeline and subject to its commercial thresholds.

The mapping is where the actual effort sits, and it is heavier here than on a simpler tool because an enterprise design system has more rules to satisfy. Somebody decides which field is the headline, what happens when a photograph is missing, how a payment figure is expressed and legally qualified, what a long trim name does to a layout, and which approval step a rendered price has to clear. Those are decisions with disclosure consequences, not switches.

Measure production rather than performance. The readable outcomes are how quickly a new offer reaches every market at every size, how much of the matrix is current rather than stale, and how many approval cycles it took. Whether the advertising then worked is a question about the media you bought, the offer you made and the speed of the follow up. Nobody can promise a result from better artwork, and this page will not.

There is no connection, and this entry carries a second gap the media platforms do not, so both are stated.

We have never made a call to this platform. No adapter file, no credential, no verification, no render job, and nothing partially written.

Connecting a production platform does not buy media. It renders files. Those files still have to be placed into networks that sell impressions, on separate accounts with separate invoices, and any pitch that blurs the two is worth a second reading.

We do not emit a vehicle catalog or an inventory feed to any advertising platform. This is the honest limitation on the dynamic creative idea. Variants could be rendered from vehicle data we hold, which is real, but that is not a live catalog handed to a network, so currency depends on when the render job last ran rather than on a continuous handoff.

We do not run dynamic creative decisioning at serve time. Choosing which variant a specific viewer sees at the moment of the impression happens inside an ad server, and we operate none.

We do not run a server side conversions API, we do not upload offline conversions, and we do not install or manage pixels.

We do not export a hashed customer list for audience matching or suppression, so audience driven variant selection is not something we could feed.

Ask by name, attach the group, and describe the matrix rather than the vendor. How many rooftops, how many manufacturer brands, how many offers a month, and who owns the design system today. Requests are recorded against the platform they name and the tally decides build order, and a request with that detail in it is worth several without.

Say as well whether you already hold a relationship with the vendor. An enterprise gate you have cleared removes the part of the timeline nobody on our side can influence, and it changes the request from a long shot into a scheduling question.

Meanwhile the managed path runs today, creative included, and waits on none of this. Platform pricing sits on the pricing page and does not vary with the services you connect.

Does LeadLocate integrate with Celtra today?

No. The status is not started, and its test is that no adapter file exists in the ad gateway for this platform. There is no credential, no checklist and no partial code, and the request count decides whether that changes.

Can I use my own Celtra account?

Not through the CRM. Nothing has been written on our side, and access is partner gated, so credentials come through an application the vendor reviews against its own commercial thresholds. An existing enterprise agreement removes the vendor half of that answer only.

What happens if I request Celtra?

It is recorded against this platform by name and counted, and the count feeds build order across the catalogue. A person answers with a range or a reason. Requests from groups that describe their creative matrix carry far more weight than requests naming the vendor alone.

Would it render advertisements from my live inventory?

Not as a live handoff. Variants could be rendered from vehicle data we already hold, but we emit no catalog feed to any advertising platform, so currency depends on when the render job last ran. A unit that sold on Tuesday can persist in rendered files until they are rebuilt.

Is this suitable for a single rooftop?

Rarely. It is enterprise tooling that expects an owner for the design system, a clean data source and a commercial commitment a single store will not clear. Without that owner you end up paying for a template nobody maintains.

Does connecting this get my advertisements placed anywhere?

No. It produces the creative. Buying the impressions is a separate decision on a separate account, and the two things arrive on different invoices. Holding that line is the most useful thing to take from this page.

Requests are recorded against the platform they name and the tally sets build order. Tell us your rooftops, your brands, your monthly offers, and who owns the design system.