Your CRM has 2,847 contacts. Your sales team swears they've contacted every prospect in your market. Yet when you run a campaign, 40% of your outreach bounces, duplicates slip through, and your team sends two emails to the same person on the same day. The problem isn't the tool—it's the data inside it. Most SMBs import contacts from at least four sources: Calendly (meeting attendees), Google Forms (landing page signups), email forwarding (manual list adds), and LinkedIn (sales team scraping). None of these sources talk to each other. A prospect books a demo on Calendly (contact created), signs up on your website (duplicate), gets added via email by a sales rep (another duplicate), and appears in a LinkedIn export (duplicate number three). Your CRM now has four records for one person. The cost is higher than you think: wasted outreach spend on undeliverable addresses, broken sales forecasts when the same deal appears twice, confused customer context when notes are scattered across four records, and lost trust when a prospect gets the same email twice in three hours. Here's how to audit your contacts, find the duplicates, and merge them back into a single source of truth. Why 40% duplicates is normal (and how to spot yours) A 40% duplication rate sounds catastrophic until you realise how contacts actually flow into a CRM. Most SMBs never set up deduplication rules. They import, then import again, and neither system knows whether "John Smith at Acme Corp" in import one is the same person as "John S, Acme" in import two. Duplicates fall into three categories: Exact duplicates: Same name, same email, created on different dates. Usually import accidents or manual adds that missed a search. Fuzzy duplicates: "John Smith" and "J. Smith", or "john.smith@acme.com" and "jsmith@acme.co.uk". Same person, slightly different data. Cross-source duplicates: The same prospect imported from Calendly (first name only, no email) and LinkedIn (full name, company, email). Your CRM sees two records. To spot yours: Sort by email domain. Export your contacts and sort by domain (@acme.com, @gmail.com, etc.). Scan for identical email addresses entered twice. A simple spreadsheet sort catches exact duplicates in seconds. Search for common names. "John Smith", "Michael Johnson"—names that appear 3+ times in your database are suspicious. One real John Smith per company is normal; three Smiths at Acme Corp is probably duplicates. Check creation dates. If two identical records were created within hours of each other, one is likely a duplicate import. Run a dedup report in your CRM. If your CRM has built-in duplicate detection (many do, including Orin's CRM ), enable it. It flags likely duplicates so you don't have to manually hunt them. Most CRMs can detect exact duplicates automatically. The real work is deciding which data to keep when two partial records exist for the same person. Audit: how to quantify the damage Before you start merging, measure the problem. You need three numbers: total contacts, likely duplicates, and impact on your pipeline. Step 1: Export and count. Export all your contacts with these fields: first name, last name, email, phone, company, date created. Open in a spreadsheet. Step 2: Identify exact duplicates. In Excel or Google Sheets: Sort by email address. Use conditional formatting (or a simple formula) to highlight duplicate emails. Count unique emails versus total rows. If you have 2,847 rows but only 1,850 unique emails, you have 997 duplicate records (35% duplication). Step 3: Identify fuzzy duplicates. Sort by last name + first name. Scan for variations: "john.smith@acme.com" and "johnsmith@acme.com", or "Michael Chen" and "M. Chen". These require manual review, but a quick scan of the exported list catches the worst offenders in 30 minutes. Step 4: Check pipeline impact. Run a report of your pipeline by contact. If the same deal appears twice (under two records for the same person), you're inflating your forecast. Count how many duplicate records have open deals attached. That's revenue you're double-counting. Example: You think you have $500k in pipeline. You actually have $280k; the rest is duplicated deals on duplicate contacts. The merge playbook: consolidating without losing data Merging contacts is not just deleting the older record. If you do that, you lose notes, activity history, and deal associations attached to the record you deleted. Before you merge, decide: which record is the master? The master record should be the one with: The most recent activity (opened emails, clicked links, attended meetings). The most complete data (email + phone + company + title, not just name). The most deal associations (if one record has an open deal, that's your master). If two records are equally incomplete, pick the newer one (more likely to have correct, current information). Step 1: Consolidate notes and history before merging. If record A has notes from three months ago and record B has an email activity fr