Latest / Marketing Automation with Fexingo: HubSpot, Marketo, and Automated Customer Journeys / Why Your Marketing Automation Needs a Data Enrichment Strategy
Transcript
- Lucas: So you spend weeks building a beautiful five-step automation sequence. Perfect trigger logic, tailored copy, personalized offers. And it lands with a thud because the person on the other end is no longer at the company, or their job title changed eighteen months ago, or the industry segment you're targeting is wrong. Luna: Ah, the classic garbage in, garbage out problem. You can have the most elegant automation in the world, but if the underlying data is stale or incomplete, you're just sending well-designed messages to the wrong people. Lucas: Exactly. And this is where data enrichment comes in. Not as a one-time cleanup project, but as an ongoing operational layer that keeps your CRM and automation platforms fed with accurate, current information. The marketing automation industry talks a lot about personalization and segmentation, but we don't talk enough about the raw material those depend on. Luna: That rings true. I mean, if this episode has sparked something you've actually used in your own workflows, the way these stay ad-free is listener support — buy me a coffee dot com slash fexingo. Lucas: Yeah, that's a nice way to think about it. We keep the episodes focused on what actually moves the needle, and listeners who find that valuable can help keep it going. So, back to enrichment — let me give you a concrete case. Lucas: I was talking to a B2B SaaS company last quarter — mid-market, about two hundred employees. They had a HubSpot instance with over forty thousand contacts, but only about thirty percent had a complete company name, industry, and revenue field. Their lead scoring model was effectively useless because the data wasn't there to score on. Luna: So they were basically flying blind on ninety percent of their database. Lucas: Right. They decided to run a batch enrichment through a third-party service that appends firmographic data — company size, industry, location, technology stack. They also layered in intent signals from content consumption. The result: within sixty days, their lead to opportunity conversion rate jumped thirty-four percent. Luna: That's massive. But I'm guessing they didn't just run one batch and call it done. Lucas: No, and that's the key. They set up a recurring enrichment cadence — every new lead gets enriched within an hour of entry, and existing records get refreshed quarterly if they show activity. They also built a data freshness SLA into their automation: if a record hasn't been enriched in ninety days, it gets flagged for re-enrichment before entering any campaign. Luna: So enrichment becomes a gatekeeper, not just a cleanup project. Lucas: Exactly. And the type of enrichment matters. There's deterministic enrichment — matching against known, verified data sources like Dun & Bradstreet or ZoomInfo. That's great for firmographic data. Then there's probabilistic enrichment, which uses statistical models to infer missing fields based on patterns. That's more common for technographic data — what software a company uses. Luna: I've seen teams get tripped up on the privacy side though. Especially with GDPR and CCPA, you can't just append data without thinking about consent. Lucas: Absolutely critical point. Most enrichment services allow you to filter by consent status. You should never enrich a record where the contact has opted out of data sharing. And you need to document your lawful basis for enrichment — typically legitimate interest for B2B, but you still need to provide a clear opt-out mechanism. Luna: What about over-enrichment? I've seen databases where every field is filled with inferred data that's only sixty percent accurate. That can be worse than having blanks. Lucas: That's the hidden cost. If you blindly accept everything an API returns, you might update a correct phone number with a wrong one. Best practice is to only enrich fields that are empty or haven't been updated in a defined period, and to use confidence scores. Most enrichment providers give a score from zero to one hundred. Set a threshold — I'd say eighty or above — and only apply those updates automatically. Below that, flag for manual review. Luna: So you're building a tiered trust model for your data. And that feeds back into automation logic. For example, a high-confidence enriched record might enter a higher-priority sequence than one with low confidence. Lucas: Precisely. And this is where enrichment connects to lead scoring. If your scoring model uses fields like company revenue or employee count, but fifty percent of your records are missing those fields, your model is effectively blind on half the database. Enrichment fills those gaps, so the model can actually do its job. Luna: I'm thinking about the downstream impact on reporting too. If your pipeline reports are built on incomplete data, you're making decisions on a fiction. Enrichment gives you a clearer picture of which segments actually convert. Lucas: Exactly. One of the companies I worked with discovered that their highest-converting segment was actually mid-market tech companies with between fifty and two hundred employees — but their original data had lumped them in with enterprise because they only had the domain name, not the employee count. Enrichment revealed the real pattern. Luna: So they were sending enterprise-level messaging to mid-market prospects and wondering why engagement was low. Lucas: Classic case. And the fix wasn't a new campaign — it was fixing the data layer. Now, enrichment isn't free. Most services charge per record, and costs can add up if you're enriching your entire database monthly. But the ROI case is usually strong. That same company calculated that the revenue from the additional conversions in the first quarter alone covered five years of enrichment costs. Luna: That's a no-brainer ROI. But I think a lot of teams hesitate because they don't know where to start. What's the first step for someone listening who thinks their data might need enrichment? Lucas: First, audit your most important fields. Pick the three fields that your lead scoring or routing depends on most. For B2B, that's often company size, industry, and job function. Run a report showing what percentage of your active leads have those fields populated. If it's below seventy percent, you have a problem. Luna: And by active leads, you mean records that have been engaged in the last ninety days, not the entire database. Lucas: Right. Don't bother enriching a five-year-old contact that never opened an email. Start with your in-market prospects. Then, pick an enrichment provider that integrates directly with your marketing automation platform. Most major platforms — HubSpot, Marketo, Salesforce Marketing Cloud — have native integrations with services like Clearbit, ZoomInfo, or Leadspace. Luna: And set up a test segment first. Enrich maybe five hundred records, run a campaign against them, and compare the conversion rate against a similar non-enriched segment. That gives you a clean A/B test to justify the budget. Lucas: Exactly. And don't forget to monitor your data decay rate. Even enriched data goes stale. On average, B2B data decays at about two to three percent per month. So after a year, a fully enriched database could be thirty percent stale again. That's why the cadence matters. Luna: So enrichment isn't a one-time fix. It's a continuous operational discipline, just like your automation sequences themselves. Lucas: Exactly. And when you treat it that way, your automation starts working the way you imagined it would. The sequences stop feeling like they're firing into the dark. You're actually reaching the right person, at the right company, with the right context. Luna: And that's probably the best ROI you can ask for from a marketing automation program. Lucas: Yeah, I think so. So the question for listeners is: what percentage of your active leads have complete data on the fields that matter most? If the number's below seventy, enrichment might be the highest-leverage investment you can make this quarter.