ADF XML vs ACP: A Side-by-Side Comparison for Automotive Lead Data

ADF XML and ACP (Automotive Commerce Protocol) are both open standards for moving automotive data between dealers, OEMs, and vendor systems. ADF was published in 2000 to carry simple text leads over email. ACP was published in 2026 to carry rich structured data, including AI-generated metadata, across modern systems. ADF is plaintext email-based. ACP is HTTPS-based, schema-validated, and AI-aware. If you are still on ADF in 2026, you are moving 2000-era data through 2000-era pipes while your competitors move 2026-era data through 2026-era pipes. The gap matters more every year.

I built ACP because ADF could not be extended any further without breaking it. This is the head-to-head comparison every dealer, OEM, and CRM vendor should read before signing their next integration contract.

What Is ADF XML?

ADF stands for Auto-lead Data Format. It is an open XML-based standard published in March 2000 by a consortium of thirteen automotive internet companies. The original purpose was to give dealerships a consistent shape for receiving consumer purchase requests from online lead providers like Autobytel, Cars.com, and Autoweb.

ADF defines five categories of data: prospect, vehicle, customer, vendor, and provider. It is delivered as XML embedded in plaintext email. Most automotive CRMs (VinSolutions, ELEAD, DealerSocket, and others) parse ADF email and extract the lead into the CRM database.

The 1.0 specification has not been meaningfully updated since 2000. The current state of the spec is still hosted at adfxml.info.

What Is ACP?

ACP stands for Automotive Commerce Protocol. It is an open standard I authored to replace ADF as the data exchange layer for modern automotive retail. The full specification is at ACPspec.org.

ACP defines canonical shapes for the objects that move between automotive systems: leads, vehicles, deals, customers, and service events. It is delivered over HTTPS as JSON with schema validation. Every object travels inside a CORD envelope that carries the schema reference, version, fidelity score, and audit trail.

The point of ACP is to make automotive data exchange compatible with how modern AI systems actually produce structured data. ADF cannot do that. ACP can.

ADF XML vs ACP: The Comparison

Eight dimensions. Side by side.

Year of publication. ADF: 2000. ACP: 2026.

Transport. ADF: SMTP email, plaintext, no encryption by default. ACP: HTTPS with TLS, encrypted by default.

Format. ADF: XML. ACP: JSON with optional XML representation.

Schema validation. ADF: DTD-based, optional, frequently skipped. ACP: Required, versioned, enforced at the endpoint.

Field coverage. ADF: name, contact, vehicle, basic timeframe. Roughly 20 fields. ACP: full lead object including intent score, AI tags, recommended next action, confidence signals, and audit metadata. Up to 50+ fields per object.

AI metadata. ADF: none. ACP: first-class fields for AI-generated intent scores, confidence levels, recommended actions, and draft messages.

Compliance. ADF: not GLBA-compliant by default due to plaintext transport. I cover this in Is ADF XML Still GLBA Compliant in 2026?. ACP: GLBA-compliant by default through HTTPS and audit logging.

Versioning. ADF: one version published 26 years ago. ACP: semantic versioning with backward compatibility guarantees.

Eight categories. ADF leads on zero of them. ACP leads on all eight. The gap is not opinion. It is structural.

Why Is ADF Still Used in 2026?

Three reasons.

Reason one. Inertia. Every automotive CRM and lead provider supports ADF. Switching means coordinating across multiple parties. Inertia is real. Inertia is also expensive in ways that are hard to see on a P&L.

Reason two. No clear successor until now. STAR exists but is complex and oriented to dealer-to-OEM transactions, not lead exchange. No one in the industry stepped up to write a true replacement. So ADF stayed by default. That changed when ACP was published.

Reason three. CRMs profit from the legacy. Every CRM that built ADF parsing has a sunk cost they want to amortize. They are not motivated to push their customers to a new standard. They will eventually have to. The pressure will come from GLBA enforcement and from dealers tired of losing AI metadata at every integration.

What Happens If You Stay on ADF?

Four predictable consequences.

You lose your AI investment at the wire. Your AI vendor charges you for a 50-field structured lead. ADF carries 20 fields. The other 30 die at the email boundary. Your BDC works the thin version of what you paid for. I cover this in What Happens When AI Captures a Lead but the CRM Cannot Hold It.

You stay non-compliant under GLBA 2022. Plaintext email transport of consumer financial information is not compliant. Your liability exposure increases every year you delay.

You cannot use AI metadata for routing or prioritization. The intent score, urgency signal, and recommended action all die before they reach the BDC. The agent treats every lead the same way because the AI signals never made it through.

Your data layer cannot evolve. ADF was not designed for AI. It was not designed for any of what is coming. Every workaround you build to extend ADF is technical debt against a standard that has no path forward.

What Should Dealers Do?

Three actions.

One. Audit your current ADF flows. List every lead provider, every CRM, every system that touches ADF in your stack. Note where AI-generated fields are being lost.

Two. Ask your vendors about ACP. Which ones support it. Which ones are committed to supporting it. The vendors that say no are committing you to the legacy mess forever.

Three. Start with one workflow. Pick one lead source. Switch to ACP. Measure the difference in lead fidelity. The numbers will make the case for the rest of the migration.

What Should AI Vendors Do?

Implement ACP. The vendors that support it can connect to any other ACP-compliant system without writing custom connectors. The vendors that resist are committing themselves to a maintenance burden that gets heavier every year.

This is the same dynamic that played out with HTTP versus FTP. The vendors that adopted HTTP early won the next decade. The vendors that stayed on FTP became legacy. ACP is HTTP. ADF is FTP.

What Should OEMs Do?

Adopt ACP at the OEM dealer portal level. This is the unlock for the industry. Once OEM portals speak ACP, dealer integrations stop being bespoke and start being standard. The industry-wide cost of moving data drops by an order of magnitude.

OEMs also benefit. Better data quality. Better visibility. Better AI-readiness across the dealer network. The ROI is obvious. The political will is harder. That is where this gets stuck today.

Final Thought

ADF XML is not bad. It is just 26 years old, built for a different era, and structurally incapable of carrying what modern AI produces. ACP is the answer the industry has been waiting for. The standard exists. The reference implementation is open. The question is how long the industry wants to keep paying the silent tax on its data layer before it switches.

Read the full ACP specification at ACPspec.org. Implement it. Push your vendors to implement it. The dealers and vendors that move early will compound for years. The ones that wait will pay the migration cost when the legacy stack finally collapses.

Structure before tools. Strategy before speed. Systems before scale.

The AI Operator Framework is the methodology for designing, governing, and scaling AI operations inside a business. Explore the full framework at JamilAshkar.com/framework. Start with the free course at JamilAshkar.com/course.