Dominion Enterprises

Bobby Gaudreau is Vice President of Sales & Marketing at Activator Dealer Solutions, where he also works in a General Manager capacity, leading the go-to-market strategy, growth, and commercialization for the customer data and marketing technology platform. He has built and scaled sales, marketing and revenue organizations across startups, PE-backed SaaS companies and larger technology businesses, and has been part of teams behind successful exits, including IMN to Reynolds and Reynolds and DealerRater to Cars.com. Gaudreau earned his MBA from the DAmore-McKim School of Business at Northeastern University and holds a BS from Keene State College.  

BUILDING EFFECTIVE, DATA-DRIVEN MARKETING STRATEGIES

Nearly everything marketers want to do now depends on the quality of the data underneath it.

At its core, marketing is communication. It is showing someone that you know who they are, understand the relationship, and can make the next interaction relevant. Like any relationship, that requires recognition, consistency, and trust.

For years, organizations could work around bad data. A salesperson might spot a duplicate record. A marketer could clean a list by hand. That becomes much harder as more of the customer journey is automated across channels.

Data quality is no longer just about valid email addresses or deduplication. It is about knowing who the customer actually is across systems, which records belong together, and which information to trust when sources disagree.

Bad data now affects marketing, reporting, attribution, personalization, decision-making, and, of course, this new thing called AI.

If marketing is about building trust, customer data is the foundation.

ENSURING HIGH-QUALITY DATA 

The biggest challenge is that customer data rarely lives in one place. In automotive, the CRM, DMS, website, marketing platforms, consent records, appended data, and third-party sources can all hold different versions of the same customer.

We have also made this harder on ourselves because automotive has developed looser definitions of terms like “data quality” and “CDP” than you typically see in the broader technology market. That topic probably deserves an article of its own.

A common issue is that many platforms start with the desired outcome, usually digital audiences, and work backward toward ROI. That can create real value, but if identity has not been resolved first, partial or conflicting customer records can still make their way into those audiences.

Then the problems move downstream. The CRM identifies one customer, the DMS tells a different story, and current website behavior may suggest something else. Reporting gets stitched together afterward. Attribution becomes harder to trust. Customer counts vary by report.

Eventually, confusion becomes mistrust.

That is why the order of operations matters. Bring together the DMS, CRM, and web data, resolve who the customer is, determine which attributes should be trusted, and only then activate, measure, and continuously maintain it.

CREATING A UNIFIED VIEW

Connection matters, but connection alone is not enough. A connected mess is still a mess.

The real value of a connected data platform is the ability to compare multiple sources, resolve identity, understand households, establish trusted attributes, and create a customer record that can actually be relied upon.

That requires something in the middle doing the hard work. An identity resolver has to determine whether three similar records are the same person. A household resolver has to understand who belongs together without making an exspouse or child the primary customer. Then the system has to decide which email, phone number, address, or consent status should win when sources disagree.

Without that layer, connecting systems is like combining every address book in a city while knowing some addresses are duplicated, many emails are invalid, and others may be spam traps. Everything is connected, but you still do not know what is true.

In dealership data, it is not unusual for us to begin with 30% to 40% of DMS records duplicated or missing important identity information before CRM and web data are even introduced.

Those same records ultimately drive your audiences and the reporting used to calculate ROI.

Integration connects the data. Resolution and governance make it trustworthy.

PEOPLE, PROCESSES AND TECHNOLOGY

Technology can do a lot of the heavy lifting, but the biggest challenge in retail automotive is more basic: nobody really owns the data.

Dealers give access to a long list of vendors, agencies, platforms, and technology partners without a clear operating model for who is responsible for the integrity of the customer record.  

I compare it to the refrigerator in a company break room. Everyone has access. Most people follow the rules. Someone eventually leaves a spoiled container of Chinese food in the back for three weeks, and now everyone has a problem.

That is essentially what happens with customer data. This is also why automotive is different from industries where enterprise CDPs have been very successful. Companies like Starbucks, Home Depot, or Nike generally own the underlying first-party data, the governance around it, and the rules for how it is managed. The CDP is operating inside an environment where someone already owns the foundation.

Retail automotive often does not have that same foundation.

We recognized that gap early at Activator.ai. It is why we spent so much time building and tuning our data management layer, identity resolver, and household resolver. Rather than trying to replace every agency or technology platform, our model is to provide the trusted underlying layer that can support them.

Think Gore-Tex or Intel.

The dealer does not need another disconnected application. It needs a trusted data layer underneath the applications it already uses.

Most importantly, someone has to own it.

THE FUTURE OF AI-DRIVEN MARKETING

AI may finally force dealers, agencies, and technology partners to solve the data problem once and for all.

AI does not fix a bad understanding of the customer. It operationalizes it and amplifies it at a scale humans never could.

Today, one employee might make one bad decision because of an incorrect customer record. Tomorrow, AI can make that same assumption across thousands of customers, choosing offers, prioritizing leads, personalizing communications, suppressing customers, and triggering workflows before anyone realizes the foundation was wrong.

That is why the sequence matters. Find the data. Group it correctly. Resolve it. Determine what is trustworthy. Maintain it as it changes. Then activate it.

We have all heard stories about someone blindly following GPS directions into a lake.

Now imagine the GPS is controlling every car on the highway.

That is the AI risk.

The companies that win will not simply have the best AI models. They will have the most trustworthy understanding of the customer underneath them.

AI is the accelerator. Data quality is the map and the steering.

If those are wrong, going faster does not make you more innovative. It just gets you lost faster.