TL;DR
Score your CRM across 7 dimensions -- completeness, accuracy, freshness, consistency, uniqueness, validity, and enrichment coverage -- using the weighted scorecard below. Run a baseline audit in under two hours, identify which dimensions are dragging your pipeline down, and fix the highest-impact issues first. Most B2B CRMs score 45-60 out of 100 on their first audit.
You cannot fix what you cannot measure. That is the core problem with CRM data quality -- most teams know their data is bad, but they cannot tell you how bad, where the worst gaps are, or what to fix first.
They guess. They run a deduplication job every quarter, clean up a few hundred records, and call it done. Meanwhile, B2B data decays at roughly 30% per year and the underlying problems keep compounding.
A proper data quality audit replaces guesswork with a score. It gives you a repeatable framework, a baseline to measure against, and a priority list that focuses effort where it matters most.
This guide walks you through the full process.
The 7 Dimensions of Data Quality
Data quality is not a single metric. It breaks into seven distinct dimensions, each measuring a different aspect of your CRM health. Scoring all seven gives you a complete picture -- and reveals problems that single-metric approaches miss entirely.
1. Completeness
Completeness measures the percentage of required fields that are actually populated across your database. A contact record with just a name and email is technically a record, but it is not actionable for outbound sales or segmentation.
What to measure: For each required field (email, phone, company, title, industry, location), calculate the fill rate across all records. Average them for your completeness score.
Why it matters: Incomplete records break lead scoring models, prevent accurate routing, and force reps to waste hours on manual research.
2. Accuracy
Accuracy measures whether the data in your CRM is actually correct. A filled field is worthless if the information is wrong -- an old email address, a job title from two years ago, or a company name for a firm that was acquired.
What to measure: Sample 200-500 records and verify key fields against LinkedIn, company websites, or an enrichment provider. Calculate the percentage that match reality.
Why it matters: Gartner estimates that poor data accuracy costs organizations an average of $12.9 million per year. Inaccurate contact data wastes outreach, damages sender reputation, and erodes trust with prospects.
3. Freshness
Freshness measures how recently your records were verified or updated. In B2B, people change jobs every 2.8 years on average. Phone numbers change. Companies get acquired. Data that was accurate six months ago may not be accurate today.
What to measure: Calculate the percentage of records updated within the last 90 days, 180 days, and 365 days. Records not touched in over a year are almost certainly stale.
Why it matters: Stale records inflate your "total addressable contacts" while delivering nothing. Outreach to outdated contacts wastes rep time and increases bounce rates.
4. Consistency
Consistency measures whether the same type of data follows the same format across your entire database. "United States" vs "US" vs "USA." "VP Sales" vs "Vice President of Sales" vs "VP, Sales." These variations look minor but break segmentation, routing, and reporting at scale.
What to measure: For fields like country, state, job title, and industry, count the number of unique values and compare against your standardized list. A country field with 47 variations of "United States" has a consistency problem.
Why it matters: Inconsistent formatting breaks every automated system built on top of that data -- territory assignment, lead scoring, marketing segmentation, and pipeline reporting.
5. Uniqueness
Uniqueness measures the percentage of records that are truly distinct. Duplicate records are one of the most damaging data quality issues in any CRM. Research from Salesforce indicates the average CRM carries a 10-30% duplicate rate.
What to measure: Run duplicate detection matching on email, name + company, phone number, and LinkedIn URL. Calculate duplicates as a percentage of total records.
Why it matters: Duplicates inflate pipeline, split activity history across multiple records, and cause reps to unknowingly work the same prospect. Every downstream metric becomes unreliable.
6. Validity
Validity measures whether data conforms to the correct format and is deliverable. An email address might be filled in (complete) and consistently formatted (consistent), but if it bounces, it is not valid.
What to measure: Run email verification across your database. Check phone numbers for correct format and connectivity. Validate URLs, postal codes, and other structured fields.
Why it matters: Email bounce rates above 2% actively damage your sender reputation. ISPs start filtering your domain, reducing deliverability on every future campaign -- not just the ones with bad addresses.
7. Enrichment Coverage
Enrichment coverage measures what percentage of your records have been augmented with additional data points beyond what was originally captured. Raw form submissions rarely contain enough data for sophisticated outbound.
What to measure: Calculate the percentage of records that have been enriched with firmographic data (company size, revenue, industry), technographic data, and verified contact information from external sources.
Why it matters: Enriched records convert at higher rates because they enable better targeting and personalization. Unenriched records limit what your sales and marketing teams can do with each contact.
Step-by-Step: Run Your First Data Audit
A CRM data quality audit follows six steps: export your data, measure completeness, check for duplicates, validate emails, assess freshness, and score the results. The entire process takes 1-2 hours for databases under 100,000 records.
You do not need expensive tools to run your first audit. A CRM export, a spreadsheet, and the scoring framework below will get you a reliable baseline.
Step 1: Export your CRM data
Pull a full export of your contact and account records. Include all fields you consider critical: email, phone, company name, job title, industry, location, last activity date, and creation date.
If your database exceeds 100,000 records, pull a representative sample of 10,000-20,000 records instead. Stratify by lead source and creation date to avoid bias.
Step 2: Run the completeness check
For each required field, calculate the fill rate:
Fill Rate = (Records with field populated / Total records) x 100
Create a table tracking fill rates for each field. Fields below 70% need immediate attention. Fields below 50% are critical gaps.
Watch For False Completeness
A field filled with placeholder values like "N/A," "unknown," or "test" is not truly complete. Filter these out before calculating your fill rate. In most CRMs, 5-10% of "filled" fields contain placeholder data.
Step 3: Check for duplicates
Run duplicate detection across three match types:
- Exact email match -- highest confidence, merge immediately
- Fuzzy name + company match -- review manually before merging
- Phone number match -- check for shared main lines vs true duplicates
Calculate your duplicate rate:
Duplicate Rate = (Duplicate records / Total records) x 100
A rate above 10% means deduplication should be your first remediation priority.
Step 4: Validate emails
Run your email list through a verification service. Categorize results as valid, invalid, risky, or unknown.
Your email validity rate is the single metric most likely to impact deliverability and campaign performance. Anything below 90% is actively hurting your outbound.
Step 5: Measure freshness
Calculate the age distribution of your records using the last modified date:
- Updated in last 90 days: __%
- Updated 90-180 days ago: __%
- Updated 180-365 days ago: __%
- Not updated in 12+ months: __%
Records in the 12+ month bucket should be flagged for re-enrichment or archival. With B2B data decaying at ~30% per year, these records have a high probability of containing outdated information.
Step 6: Score the results
Use the scoring framework in the next section to convert your raw metrics into a weighted score out of 100. This becomes your baseline -- the number you improve against over time.
The Data Quality Scorecard
This is the framework that turns raw metrics into an actionable score. Each dimension gets a weight reflecting its impact on revenue and operations. Score each dimension 1-5, multiply by the weight, and sum for your total.
| Dimension | Weight | Score 1 (Critical) | Score 2 (Poor) | Score 3 (Average) | Score 4 (Good) | Score 5 (Excellent) |
|---|---|---|---|---|---|---|
| Completeness | 20% | Under 50% fill rate | 50-65% | 65-80% | 80-90% | Over 90% |
| Accuracy | 20% | Under 60% verified correct | 60-72% | 72-84% | 84-93% | Over 93% |
| Freshness | 15% | Under 30% updated in 90 days | 30-50% | 50-65% | 65-80% | Over 80% |
| Consistency | 10% | Over 40% format variations | 25-40% | 15-25% | 5-15% | Under 5% |
| Uniqueness | 15% | Over 25% duplicates | 15-25% | 8-15% | 3-8% | Under 3% |
| Validity | 15% | Under 80% valid emails | 80-88% | 88-93% | 93-97% | Over 97% |
| Enrichment Coverage | 5% | Under 20% enriched | 20-40% | 40-60% | 60-80% | Over 80% |
How to calculate your total score:
Total = (Completeness Score x 0.20) + (Accuracy Score x 0.20)
+ (Freshness Score x 0.15) + (Consistency Score x 0.10)
+ (Uniqueness Score x 0.15) + (Validity Score x 0.15)
+ (Enrichment Coverage Score x 0.05)
Maximum possible = 5.0 (then multiply by 20 for a 0-100 scale)
Score interpretation:
- 80-100: Excellent. Your data is a competitive advantage. Focus on maintaining.
- 60-79: Good. You have solid foundations with specific areas to improve.
- 40-59: Average. Data quality is limiting your sales and marketing effectiveness.
- 20-39: Poor. Data problems are costing you pipeline and productivity daily.
- Under 20: Critical. Stop outbound campaigns until you address core data issues.
Benchmark
Most B2B companies score 45-60 on their first audit. If you score above 70 without prior data quality investment, your team has been doing something right. If you score below 40, you are leaving significant revenue on the table.
B2B Data Quality Benchmarks
How does your data compare to the industry? These benchmarks are drawn from Gartner, Experian, Salesforce, and Dun & Bradstreet research across B2B companies.
| Metric | Poor (Bottom 25%) | Average (Median) | Good (Top 25%) | Excellent (Top 10%) |
|---|---|---|---|---|
| Email accuracy | Under 82% | 82-90% | 90-95% | Over 95% |
| Phone fill rate | Under 35% | 35-55% | 55-75% | Over 75% |
| Company data completeness | Under 55% | 55-72% | 72-85% | Over 85% |
| Duplicate rate | Over 20% | 10-20% | 5-10% | Under 5% |
| Data freshness (updated in 6 months) | Under 40% | 40-60% | 60-78% | Over 78% |
A few things stand out in these numbers.
Phone fill rate is universally low. Even top-performing companies only reach 75%+ direct dial coverage. This is because phone data is harder to source and verify than email. Waterfall enrichment helps by querying multiple providers for the same record, increasing the odds of finding a verified direct dial.
Duplicate rates are higher than people assume. The median B2B CRM carries 10-20% duplicates. Most teams underestimate this because duplicates hide across objects -- a contact duplicate and an account duplicate for the same person.
Data freshness degrades fast. If you are not actively re-enriching records, your freshness score drops by roughly 2-3 percentage points per month. Within a year, nearly a third of your database is outdated.
What to Fix First: Priority Matrix
Not all data quality problems deserve equal attention. Use this matrix to prioritize based on impact and effort.
| Easy to Fix | Hard to Fix | |
|---|---|---|
| High Impact | Email validation, deduplication, lead source tracking | Accuracy verification, company hierarchy cleanup |
| Low Impact | Phone formatting, address standardization | Full re-enrichment of archived records, historical data backfill |
Start in the top-left quadrant. Email validation and deduplication are high-impact and straightforward to execute:
-
Email validation can be run across your entire database in hours. Removing invalid addresses immediately improves deliverability and bounce rates. Use Cleanlist email verification to catch invalids before they damage your sender reputation.
-
Deduplication has clear matching rules (email match, name + company match) and most CRMs have built-in detection. Merging duplicates instantly improves pipeline accuracy and prevents reps from working the same prospect.
-
Lead source tracking is a process fix -- enforce required fields at the point of entry. No tool cost, no data migration, just configuration.
Move to the top-right next. Accuracy verification and company hierarchy cleanup take more effort but have significant downstream impact on lead scoring, routing, and attribution.
Avoid the bottom-right until everything else is done. Full historical backfill and archived record re-enrichment have diminishing returns. Those resources are better spent on maintaining quality for active records.
Automate Ongoing Quality Monitoring
A one-time audit tells you where you stand. Ongoing monitoring keeps you there. Without automated quality checks, data decays back to its pre-audit state within 6-12 months.
Set up automated triggers
Configure your CRM or data tools to enforce quality on every new record:
- On lead creation: Verify email, enrich missing fields, check for existing duplicates
- On email bounce: Flag the record, trigger re-enrichment, suppress from future campaigns
- On field update: Validate format consistency (phone, country, state)
Schedule recurring audits
| Audit Type | Frequency | What to Check |
|---|---|---|
| Duplicate scan | Weekly | New duplicates created since last scan |
| Email validity check | Monthly | Re-verify emails on active outreach lists |
| Completeness report | Monthly | Field fill rates across required fields |
| Full 7-dimension audit | Quarterly | Run the complete scorecard, compare to baseline |
| Re-enrichment pass | Quarterly | Refresh records not updated in 90+ days |
Build a quality dashboard
Track three metrics weekly at minimum:
- Overall data quality score (from the scorecard framework above)
- New duplicate creation rate -- measures whether prevention is working
- Email bounce rate trend -- the earliest signal of data decay
When your score drops more than 5 points between quarterly audits, investigate the root cause before it compounds. The most common culprits are bulk imports without deduplication, new lead sources with lower data quality, or a lapsed re-enrichment schedule.
Practical Starting Point
If setting up full automation feels overwhelming, start with one rule: verify every email address before it enters your CRM. This single check prevents the highest-impact data quality problem -- invalid emails damaging your sender reputation. Build from there.
Frequently Asked Questions
How often should I audit CRM data?
Run a full 7-dimension audit quarterly. Monthly spot checks on the three highest-impact metrics -- email validity, duplicate rate, and field completeness -- catch problems before they compound. High-velocity sales teams running aggressive outbound should check email validity weekly. The key is consistency. A quarterly audit that actually happens is better than a monthly audit that gets skipped.
What's a good data quality score?
A score of 60-79 out of 100 puts you in "Good" territory -- ahead of most B2B companies. Most organizations score 45-60 on their first audit. Anything above 80 is excellent and indicates your data is a genuine competitive advantage. If you score below 40, data quality issues are actively costing you pipeline and should be treated as urgent. Use the CRM data quality benchmarks post for detailed metric-level targets.
How much does bad data actually cost?
Gartner estimates bad data costs organizations an average of $12.9 million per year. For SMBs, the figure is typically $200,000-$750,000 annually across bounced emails, wasted rep time, failed automations, and missed opportunities. The cost compounds because data decays at roughly 30% per year -- so doing nothing means next year is worse. We break down the full calculation in our cost of bad data framework.
Can I automate data quality checks?
Yes, and you should. Email verification on lead creation, automated duplicate detection, format validation rules, and scheduled re-enrichment can all be automated. Most teams start with 80% manual data quality work and should target 80% automated within six months. The mechanical tasks -- verification, format enforcement, duplicate flagging -- are ideal for automation. Keep human judgment for merge decisions, archival choices, and scoring model recalibration.
Your CRM data quality is either a growth lever or a growth blocker. There is no neutral state. Run the audit, score your baseline, fix the top-left quadrant of the priority matrix first, and set up recurring checks so the gains stick. If you want to automate the heaviest parts -- email verification, enrichment, and ongoing monitoring -- see what Cleanlist can do for your data.