Latest / Marketing Automation with Fexingo: HubSpot, Marketo, and Automated Customer Journeys / How Marketing Automation Workflows Hide Data Rot
Transcript
- Lucas: I found a number this week that genuinely unsettled me. According to a study from Dun & Bradstreet, contact data decays at about twenty-three percent per year. That means if your marketing automation database has ten thousand records, by next June roughly two thousand three hundred of them are wrong. Luna: Twenty-three percent. That's almost a quarter. And wrong how — like wrong email addresses, or wrong job titles, or both? Lucas: Both. The study breaks it down: email addresses bounce at about two point one percent per month, phone numbers go bad at roughly three percent per month, and job titles — especially in B2B — shift about fifteen percent annually. So the rot is cumulative and it hits different fields at different speeds. Luna: Right, and if you're running a nurture sequence that depends on a contact's title or industry code, a fifteen percent annual drift means your segmentation is quietly falling apart. Lucas: Exactly. I talked to a marketing ops lead at a mid-market SaaS company — they had a workflow that scored leads based on job seniority. VP and above got a higher score and a faster sales handoff. But when they audited the data, twenty percent of the 'VP' titles were actually people who had left those roles a year ago. Their automation was systematically over-heating leads that were cold. Luna: So sales was chasing ghosts. Did they catch it early? Lucas: Not early enough — they had been running that workflow for fourteen months. But once they cleaned the data and added a 'date of last verification' field to every contact record, they saw a twelve percent lift in meeting booking rates within sixty days. The fix wasn't a new tool or a new channel — it was simply knowing which data was old. Luna: That's the part that surprises me. Most teams obsess over trigger logic — what email to send, when to send it, how many steps in the flow. But they rarely ask: is the data feeding this trigger even accurate? Lucas: Right. And the problem is invisible until something breaks. A nurture sequence that used to convert at three percent drops to one point five percent — and the knee-jerk reaction is to rewrite the email copy or change the send time. But the real issue is that half the contacts in the sequence no longer work at the companies the system thinks they do. Luna: I want to get practical here. What's the cheapest way to start auditing for data rot? Lucas: The simplest starting point is a bounce-rate report. Pull the last three months of email sends from your platform — HubSpot, Marketo, whatever you use. Segment by campaign or workflow. If any single workflow has a hard bounce rate above three percent, that's a signal that the data feeding it is stale. Then you can look at the source of those contacts: did they come from an old import? A third-party list? A form that hasn't been updated in two years? Luna: Three percent — that's a good benchmark. And for B2B, I'd add: check the domain part of the email. If you see a lot of @aol.com or @oldcompany.com, those people have probably moved on. Lucas: Great point. Another easy win is to run a 'title freshness' check. Export all contacts with a title field that hasn't been updated in twelve months. Then use a free tool like LinkedIn Sales Navigator to spot-check a random sample of fifty. You'll usually find that ten to fifteen percent are wrong. That alone justifies a quarterly data refresh. Luna: And that refresh doesn't have to be manual. Most automation platforms let you set up a periodic data append or verification workflow. You can trigger a re-verification every ninety days for any contact that hasn't been updated in a year. Lucas: Exactly. And the ROI is almost immediate. If you have a hundred thousand contacts and only fifty thousand are reachable, you're paying for storage on fifty thousand dead records. Plus you're sending emails that hurt your sender reputation. Cleaning the rot improves deliverability, which improves engagement, which improves the whole flywheel. Luna: There's another angle I want to throw in: data rot doesn't just break automation — it breaks analytics. If your pipeline reports are based on contact records that are twenty percent stale, your forecasts are wrong. Your attribution is wrong. You might be making budget decisions based on fiction. Lucas: That's the hidden cost. The marketing team I mentioned earlier had been reporting a certain number of 'active leads' to the board every quarter. When they cleaned the database, that number dropped by eighteen percent overnight. That was scary — but it was accurate. And it forced better upstream data collection. Luna: Speaking of accurate: if this episode has ever made you look twice at your own database, or saved you from sending one more email to someone who left their job two years ago — honestly, if that was worth a coffee to you, there's a link. It's buy me a coffee dot com slash fexingo. No pressure, just a way to keep these conversations going. Lucas: Yeah, and it's entirely listener-supported — no advertisers, no sponsors. So if you've gotten value from the show, that's the place. And we'll keep digging into the mechanics of automation that actually works. Luna: So back to the practical: if a team wants to start a quarterly data hygiene program this month, what's the first thing they should do? Lucas: Schedule a one-hour block to run a bounce-rate and title-freshness report. Then set up a recurring task to do the same thing every quarter. And add a custom field called 'date of last verification' on every contact record. That single field makes the invisible rot visible. Within two cycles, you'll have a clean database that your automation can actually trust. Luna: And that's the foundation. No amount of clever workflow logic can save you from bad data. Lucas: Exactly. Clean data first, smart automation second. Always.