Confessions of a Dirty Field

by Sue Wieberg

Global Head of Delivery

Sojourn Solutions

The rise and fall of the contact field, {misc_info_3} in the age of AI and ABM

One day I held important data that supported the Marketing Ops team with critical data that facilitated personalization for an event campaign. 

The next? “Loves pickles.”

I’ve seen emojis, lorem ipsum, broken HTML, and once - a confession about someone’s boss.

I am full of secrets.

And none of them are good.

I wasn’t always like this. I was created with purpose. Potential. They said I’d be “a flexible metadata field for edge cases.” The kind of field you use when the structured fields just don’t cut it. But somewhere along the way, I became the junk drawer of your Marketing Automation Platform.

Now, in the age of AI and account-based marketing, I sit silently in the shadows… bloated, unindexed, and untrustworthy. I know I shouldn’t be here, but no one has the heart to delete me.

The high stakes of dirty data in modern Marketing Ops

The analysts avoid me. The business analytics team refers to me as ‘that cursed

field.’ I once made a data scientist cry. They tried to parse me. They failed.

You’re running AI-generated nurture sequences and building ABM tiers based on firmographics and behavioral intent. But I have to ask…are you really ready for automation when I am still in your database? Alone and unclean?

Data like mine - messy, unlabeled, ambiguous - kills AI’s potential. And ABM? It’s nothing without clean, actionable data. Dirty fields like me lead to: Confused targeting; Awkward personalization; Broken integrations; Wasted spend; and Deep existential shame (mostly mine).

What I wish you’d done instead

I didn’t want this life. You gave it to me. With every import, every lazy copy-paste,

every ‘just for now,’ you chipped away at my purpose.

Here’s what your modern Marketing Ops data strategy should be built on

1. Clear Purpose for Every Field

If a field exists, it should have a name, a defined use, and a reason to live. I never got that. I was your “misc” field. Your fallback. Your data orphan.

2. Input Validation & Guardrails

People shouldn’t be able to paste 1,000-character XML into a text field meant for "preferred language." But they did. To me. Daily.

3. Field Audits & Schema Reviews

Look back. What am I even doing here? When was the last time someone reviewed the fields you're actually using? Spoiler: it's not the ones you think.

4. Structured Data for AI & ABM

Machine learning needs consistency. ABM needs segmentation. I provide neither. If your programs rely on me, they’re doomed.

5. Documentation & Training

Teach your teams how to treat their data with care. Show them the horror that I’ve become and let me be a cautionary tale.

AI and ABM is not a Band-Aid for broken data

Sometimes I fantasize about being renamed. Something dignified like {aux_metadata} or {custom_notes_json}.

But no.

I am, and will always be {misc_info_3}.

Forgotten.

Feared.

Full of things no one should ever see.

We’re in a new era of Marketing Ops. AI is powerful. ABM is precise. But neither of them can thrive on a foundation made of random fields, duplicated notes, or wild data like mine. If you want to automate smarter, target better, and move faster, you have to fix your foundation. Start with your fields. Rescue the ones you can. Retire the ones like me.

Let me go.

Please.

Final word: From {misc_info_3} and your future AI model

You didn’t think I mattered.

But now, I’m the reason your AI personalization is suggesting cat food to a CFO.

Good luck explaining that in your QBR.

Don't wait until your strategy breaks to realize your data is the problem. Clean data isn’t a “nice to have.” It’s the difference between a smart, scalable marketing engine - and a chaotic mess with a dash of existential dread.

Need help turning your data landfill into a growth engine? We specialize in modern Marketing Ops: cleaning data, building smart schemas, and preparing your tech stack for the future.

Let's finally give {misc_info_3} the closure they deserve.

About the Author