What is data quality?
Data quality is the measure of how well data serves its intended purpose. In B2B systems, data quality is evaluated across several dimensions — accuracy, completeness, timeliness, consistency, and uniqueness. Poor data quality costs organizations an estimated $12.9 million per year (Gartner) and undermines every downstream process from sales outreach to compliance reporting.
Data quality is not a one-time project — it requires continuous monitoring and maintenance. CRM data decays at 30-40% per year as people change jobs, companies restructure, and contact information goes stale.
Dimensions of data quality
- Accuracy — Is it correct? Does the job title in your CRM match what the person actually does today? Is the company revenue figure current or from three years ago?
- Completeness — Are all fields populated? A record with a name and email but no title, company, or phone number is incomplete and limits segmentation, routing, and outreach.
- Timeliness — Is it current? A record that was accurate six months ago may already be stale if the person changed jobs, the company restructured, or the phone number changed.
- Consistency — Do systems agree? If your CRM says "Salesforce" and your marketing platform says "salesforce.com, inc.", the same entity appears as two different companies in reports.
- Uniqueness — Are there duplicates? Duplicate records inflate pipeline metrics, break routing rules, and cause reps to contact the same person multiple times.
How to measure data quality
Field fill rates measure completeness — what percentage of records have key fields like title, direct phone, and company populated. A CRM with 60% title fill rate has a significant enrichment gap.
Bounce rates are a lagging indicator of email accuracy. If your email bounce rate is above 5%, your contact data is materially stale.
Duplicate rates measure uniqueness. Enterprise CRMs average 20–30% duplicate rates — meaning a fifth to a third of all records are redundant.
Decay rate measures how quickly data goes stale. B2B contact data decays at 30–40% per year, but the rate varies by industry and seniority level.
Confidence scores provide a field-level quality signal — how certain is the system that a given data point is correct? This transforms data quality from a periodic audit into a real-time dashboard.
How Salmon improves data quality
Salmon maintains data quality continuously — not through quarterly cleanups. AI agents verify every record against live sources, flag stale fields, merge duplicates, and keep your CRM current. Confidence scores on every field give your team a real-time data quality dashboard.
Why data quality degrades
Data quality is not a set-it-and-forget-it problem. Several forces continuously erode the quality of your CRM data:
- Job changes — The average tenure in B2B roles is 18–24 months. Every time someone changes jobs, their title, company, and often email become stale in your CRM.
- Company restructuring — Mergers, acquisitions, rebranding, and reorganizations change company names, hierarchies, and employee roles — often without any external signal.
- Manual entry errors — Reps type quickly. Nicknames, typos, and inconsistent formatting accumulate into a mess of duplicate and conflicting records.
- Integration drift — As systems sync, mapping errors and deduplication failures introduce inconsistencies that compound over time.
- Stale vendor data — Third-party data providers refresh their databases periodically (often quarterly), meaning the data you buy is already aging the moment you import it.
Related concepts
See real-time enrichment on your data.
Send us a sample from your CRM. We'll show you what Salmon enriches, verifies, and fixes — live, in 30 minutes.