Nobody wakes up and decides to clean their CRM. It happens when something breaks. An email campaign bounces at 18%. A rep discovers they've been calling the same prospect as three other reps. A board report shows pipeline numbers that don't match reality.
By the time these problems surface, the underlying data quality issues have usually been compounding for months. Here are the warning signs, ranked by how early they appear.
1. Your Email Bounce Rate Is Climbing
Healthy B2B email campaigns bounce at 2-3%. If you're consistently above 5%, your contact data has decayed past the point where it's reliable.
What's happening: People change jobs. Companies get acquired. Email addresses get deactivated. If you haven't validated your email list in the last 6 months, roughly 15% of your addresses are probably bad.
The fix: Run SMTP validation on your entire email database. Remove hard bounces immediately. Flag soft bounces for re-verification in 30 days. Set up automated validation on new contacts as they enter your CRM.
Related: Email Deliverability and Data Quality
2. Reps Are Complaining About Duplicates
When sales reps start mentioning that they're calling prospects who've already talked to another rep, you have a deduplication problem. This is more than an annoyance. It damages your brand and wastes selling time.
What's happening: Duplicates accumulate from trade show imports, website form submissions with slight name variations, data purchased from multiple sources, and manual entry without duplicate checking.
The fix: Run fuzzy matching across your entire database (not just exact email matches). Merge duplicates into golden records. Implement real-time duplicate detection on new record creation.
Related: What Is Data Deduplication?
3. Lead Routing Is Sending Records to the Wrong Rep
If your lead routing rules depend on fields like state, industry, or company size, and those fields are inconsistent or missing, leads will get misrouted. This is one of the most expensive data quality failures because it directly delays time-to-contact.
What's happening: State fields have a mix of abbreviations and full names ("CA" vs. "California"). Industry codes are missing or incorrect. Company size data is outdated.
The fix: Standardize the fields your routing rules depend on. Fill missing values through enrichment. Audit routing logs monthly to catch misroutes early.
4. Marketing Segments Return Wildly Different Counts Each Time
You build a segment of "VP-level contacts at companies with 100+ employees in financial services." You get 3,200 results. You rebuild the same segment next week with slightly different title filters and get 1,800.
What's happening: Job titles aren't standardized. "VP of Finance," "Vice President, Finance," "VP Finance," and "Finance VP" are four different text strings that represent the same role. Without normalization, every filter gives different results.
The fix: Normalize job titles to a consistent taxonomy. Standardize company size ranges. Clean industry codes. Once fields are consistent, segments become reliable and repeatable.
Related: Job Title Normalization
5. Your CRM Reports Don't Match Reality
The pipeline report says you have 200 opportunities. Your sales manager says it's more like 120. The revenue forecast is off by 30% every quarter.
What's happening: Duplicate deals inflate pipeline. Outdated stage values misrepresent where deals actually are. Missing close dates create phantom opportunities. Inconsistent naming makes it impossible to track accounts across stages.
The fix: This requires both data cleaning and process fixes. Clean the existing pipeline data (merge duplicates, update stages, remove dead deals). Then implement required fields and validation rules to prevent the same problems from recurring.
6. Half Your Phone Numbers Go to Voicemail or Are Disconnected
If your connect rate on outbound calls has dropped below 15-20%, your phone data is stale. Reps are dialing disconnected numbers, former employees, and main office lines instead of direct dials.
What's happening: Phone numbers decay faster than email addresses. People change numbers when they change jobs. Direct dials get reassigned. Cell numbers change carriers. Main lines get rerouted during office moves.
The fix: Validate phone numbers against carrier databases. Flag disconnected lines, landlines, and main office numbers separately from verified direct dials and mobile numbers. Re-enrich contacts where the phone data is stale.
7. New Hires Can't Find Anything in the CRM
You hire a new rep. They spend their first week trying to search the CRM for accounts in their territory. They can't find half of them because company names are entered inconsistently, territories are coded wrong, and related contacts aren't properly associated.
What's happening: Years of inconsistent data entry have created a database that only makes sense to the people who entered the records. New users can't navigate it because the data doesn't follow any predictable pattern.
The fix: Standardize company names, normalize address formats, and clean up account-contact associations. The goal is a CRM where any user can search intuitively and trust what they find.
What to Do Next
If three or more of these signs apply to your CRM, you're past the point of quick fixes. You need a systematic cleaning pass:
- Audit first. Measure your duplicate rate, bounce rate, fill rate, and standardization level. You can't fix what you haven't measured.
- Prioritize by impact. Start with the issues that directly affect revenue (email deliverability, lead routing, pipeline accuracy).
- Clean, then maintain. A one-time cleanup loses its value within 6 months without a maintenance plan.
Frequently Asked Questions
How do I know if my CRM data is bad?
The clearest indicators are email bounce rates above 5%, sales reps reporting duplicates, lead routing errors, inconsistent marketing segments, and CRM reports that don't match reality.
How often should you clean CRM data?
At minimum, quarterly. B2B data decays at 2-3% per month. A database that was clean in January will have 10-15% bad records by April.
What is the fastest way to clean CRM data?
Outsource to a managed provider (3-7 day turnaround). For in-house, start with email validation, then deduplication, then field standardization. Don't try to fix everything at once.