Your RealGreen database is a gold mine (after you clean it)

A ten-year-old RealGreen database is the most valuable marketing asset a lawn care company owns (thousands of current and former customers, their properties, their service history), and most of them are unusable as-is: duplicated, uncontactable, miscoded, unmeasured. The fix is a four-step cleanup in a specific order (status, identity, contactability, enrichment), and the payoff is every campaign that was impossible before it.

What's actually wrong with your database?

Nothing that breaks daily operations. That's exactly why it never gets fixed. Routes run and invoices post just fine on top of:

• Duplicate accounts. The customer who cancelled in 2018 and re-signed in 2022 as a "new" account; the Smiths at 114 Maple and the Smyths at 114 Maple Ave.
• Dead contact info. Emails that bounce, landlines that disconnected years ago, and (worse for texting) numbers that were never flagged mobile vs. landline.
• Status fiction. "Active" customers who haven't taken a service in three seasons; cancels with no reason code (the problem we covered in the cancel-data post); one-time aerations entered as programs.
• Measurement gaps. Property records with no turf square footage, or a number a tech eyeballed in 2016. Every quote and material estimate built on it is a guess.

None of this hurts until you try to market to the database, and then all of it hurts at once: the win-back email goes to a current customer's duplicate, the prepay letter goes to someone who moved, and the upsell is priced for the wrong lawn.

What order should you clean in?

Order matters because each step shrinks the work of the next. Validate emails before merging duplicates and you'll pay to validate the same dead address twice.

Step	What you're fixing	How	What it unlocks
1. Status	Who's genuinely active, cancelled, or one-and-done	Rule-based: no completed service in N months → recode; backfill cancel reasons where the notes say why	Honest counts; a real win-back list
2. Identity	Duplicate people and properties	Match on address + name/phone similarity; merge histories to one surviving account	One customer, one story, one suppression decision
3. Contactability	Dead emails, bad phones, mobile vs. landline	Bulk validation services; flag SMS-capable numbers; capture consent going forward	Email & SMS campaigns that actually deliver
4. Enrichment	Missing turf measurements and property data	Bulk address-based measurement (DeepLawn) written back to property records	Priced upsells and quotes without a truck roll

Steps one and two are largely one-time projects. Steps three and four need a maintenance loop (bounces flagged monthly, new customers measured at entry), or entropy wins again within two seasons.

How do you fix measurement gaps at scale?

This is the step owners assume is impossible and is actually the most automatable. Address-based measurement through DeepLawn pulls turf square footage from imagery in bulk: run the whole customer file through, write the results back to the property records, and flag the outliers (heavily wooded lots, new construction) for a human look. From then on, every renewal price, every condition-code upsell, and every neighbor quote prices off real square footage: the same measurement backbone behind instant quotes. We went deeper on the measurement economics in the property-measurement post.

What does clean data unlock for campaigns?

Everything that was quietly impossible before:

• Win-backs to a real list of properly coded former customers, the warmest cold list in existence.
• Prepay season that suppresses correctly: no discount letters to past-dues, no "final reminder" to someone who paid in November (the mechanics are in the prepay playbook).
• Upsells priced off real measurements, triggered by condition codes, sent to addresses that can actually receive them via email & SMS.
• Honest reporting. Cost per sold customer, churn rate, customer lifetime value. Every one of those numbers is fiction when the denominators are full of duplicates and phantom actives.

The plumbing that keeps it clean is the nightly sync between RealGreen and your marketing system, the workaround architecture we described in the RealGreen API post, so statuses, payments, and new codes flow through automatically instead of via October list exports.

What does the cleanup cost versus return?

Honest effort estimate: a 5,000-account database that's never been cleaned is typically a few weeks of rules-plus-review work: bulk email validation runs pennies per address, bulk measurement runs dollars per property, and the labor is mostly in the duplicate review and status rules. Call the whole project low four figures in hard cost plus the project time, illustratively. Against that: a win-back campaign to 800 properly coded former customers converting even 3–5% at a $550 average program is $13,000–$22,000 of first-year revenue, from the first campaign the cleanup makes possible. The prepay suppression fixes and deliverability gains compound quietly behind it. One compliance note while you're in there: capture and record SMS consent properly as numbers get validated. A clean database that texts people who never opted in trades a data problem for a legal one. Get consent language into your intake forms and quote flows now, and the list grows clean from here.

Who should actually do the work?

Split it by judgment required. The rule-based passes (status recodes, bounce flagging, bulk validation, bulk measurement) are scripts and services, not staff time; don't burn your office manager's March on them. The judgment calls (duplicate merges where histories conflict, backfilled cancel reasons, outlier measurements) need someone who knows the customers, in short weekly batches rather than a death-march week. And resist the temptation to "just fix it in the marketing system" while leaving RealGreen dirty: RealGreen is the system of record your office and techs live in, so a cleanup that only exists in a synced copy decays the moment someone edits the source. Clean the source, sync the copy, and both systems tell the same story.

The takeaway: clean in order (status, identity, contactability, enrichment), automate the maintenance loop, backfill measurements in bulk with DeepLawn, and your ten-year-old RealGreen database becomes the highest-ROI marketing channel you own.

Your database cleanup checklist

1. Recode status: no completed service in 18 months → inactive, with a backfilled reason where possible.
2. Merge duplicates on address + name/phone match; keep full history on the survivor.
3. Bulk-validate emails and phones; flag mobile numbers for SMS.
4. Run the file through DeepLawn; write turf square footage back to every property.
5. Stand up the maintenance loop: monthly bounce handling, measure-at-entry for new customers.
6. Launch the first unlocked campaign: the win-back list usually pays for the whole cleanup.