# HubSpot Duplicates Left by a CRM Rollout | Dedupely > A CRM consultant helped an IT staffing firm merge 16,000+ duplicate HubSpot records left by its rollout, with merge rules agreed with the client. - URL: https://dedupe.ly/case-studies/staffing-firm-consultant-hubspot-duplicates - Section: case-studies - Published: 2026-10-20 - Site index: https://dedupe.ly/llms.txt - Any page: append `.md` to its URL for this plain-text form --- [Case studies](https://dedupe.ly/case-studies) HubSpot Partner story ## How a CRM Consultant Helped a Staffing Firm Merge 16,000+ Duplicate HubSpot Records Left by Its Rollout, with Dedupely - [HubSpot](https://dedupe.ly/integrations/hubspot) - After a list import - Integration or sync - Recruiting - 16,000+ duplicate records merged - 8,200+ matches merged - 54,507 records synced to Dedupely A match is one set of duplicate records merged into a single record. Every match holds at least two records, so the number of records merged is at least double the number of matches. TL;DR An IT staffing firm's HubSpot rollout left it with duplicates of most of its companies and contacts. Its CRM consultant agreed merge rules with the firm, built searches in layers from obvious to tricky, protected company branches and merged more than 16,000 duplicate records. On this page - 01 Where duplicates came from - 02 What duplicates broke - 03 How duplicates were solved - 04 What deduplication changed ### Where duplicates came from #### A rollout that loaded everything more than once The client is an IT staffing and recruitment firm with several offices. It runs sales and marketing in HubSpot and keeps its companies and contacts in an applicant tracking system (ATS), which is its source of truth. An independent CRM consultant, who helped the firm roll out HubSpot and works as its product manager, took on the duplicate problem. The duplicates came in during the rollout: - **Big imports from the ATS.** When HubSpot went live, contacts and companies were imported from the ATS in bulk, and an integration set up at the time made things worse. - **Bad loads.** Some loads brought in company names at the wrong address or with the wrong ATS ID. - **Two web domains for one company.** Many companies have both a .com and a country domain, and HubSpot treated them as two different companies. The result was about 22,000 companies where there should have been around 10,000, and 120,000 contacts, some of them in HubSpot three, four or five times. ### What duplicates broke #### Activity logged on the wrong record Sales logged calls and notes against one record for a contact, then looked the person up again and couldn't find their own history. Merging in HubSpot meant two records at a time, with a daily cap, and the totals barely moved. #### A marketing program on hold The firm wanted to launch a marketing program, and it couldn't with so many duplicates. The same person could get the same email several times, or HubSpot would stop the send. The consultant said they simply needed to clean the data before they could run a program they could rely on. ### How duplicates were solved - #### 1. Test first, with the client signing off The firm has its own Dedupely account, and the consultant worked in it as a guest user, so the tool stays with the firm after her work ends. She tested searches on old, untouched records first, and nothing ran across the whole database until the firm's CIO was comfortable with it. - #### 2. Agree which values win, field by field Before merging, the consultant and the firm agreed what should survive a merge, field by field, and the firm signed off on the [Merge Rules](https://dedupe.ly/product/duplicate-merging): - **Primary Record:** the oldest record. - **Email and phone:** the record with the most recent activity wins, since that's most likely the email and number the person uses now. - **Bad emails lose:** Lose Rules push out the email from contacts whose status is Inactive, emails that hard bounced, and test addresses containing words like dummy, fake or example. - **The ATS ID:** the ID from the most recently created record wins, because the newest load from the ATS carries the correct ID. - **Former employees' records lose:** on email, phone, owner and contact status, values from records owned by a former employee lose, so the active team's details survive. - **Lead Status and Lifecycle Stage:** kept from the primary record. Blank values never overwrite existing data. - #### 3. Merge in layers, from obvious to tricky The first company search matched on four things at once: a similar company name, the street address (Similar Words, any order), the company owner and the ATS ID. Matches that agree on that much are good candidates for a bulk merge. Later layers loosened one field at a time, so the obvious duplicates could go first and the harder cases got more care. - #### 4. Keep company branches apart Many companies have several offices that share a name. To stop branches from merging into each other, the consultant worked through one search on its own: a similar company name and the exact ATS ID, with a filter so at least one record in each match had no address. While working this search, the Safety Settings for [Companies](https://dedupe.ly/integrations/hubspot/duplicate-companies) were set to two records per match, so anything larger was held back in Excluded instead of merging. Because Safety Settings apply to every Companies search, this search was handled by itself. That way, a record without an address merges into the one that has it, and two offices that both have an address won't be combined. The consultant ran these matches as a bulk merge. Alongside the merges, the consultant and the firm's CIO deleted about 7,200 blank company records that had no activity. ### What deduplication changed #### 16,000+ duplicate records merged The firm has merged more than 8,200 matches, combining at least 16,000 duplicate records. Today, Dedupely syncs 54,507 records for the firm. #### A cleanup the firm can keep running Early on, the consultant said she thought Dedupely was the right tool for the client. Because the account belongs to the firm, the plan is for someone on its own team to learn the rules and searches and keep the data clean after the initial cleanup. Takeaway If someone is cleaning up your CRM for you, sit down together before anything is merged and agree which email, phone and ID should win. Put those decisions into merge rules, so every merge follows them. ### Related - [Merge Rules](https://dedupe.ly/knowledge-base/merge-rules) - [Filters](https://dedupe.ly/knowledge-base/filters) - [Merge Safety](https://dedupe.ly/knowledge-base/merge-safety) - [Merge Rules](https://dedupe.ly/glossary/merge-rules) - [CRM Implementation](https://dedupe.ly/glossary/crm-implementation) - [Customer story 9,000+ duplicate records merged HubSpot Healthcare How a Telehealth Company Merged 9,000+ Duplicate HubSpot Records Created by Its Calling App, Forms and Integrations, with Dedupely Read the story](https://dedupe.ly/case-studies/telehealth-hubspot-duplicate-contacts) - [Customer story 1,600+ duplicate records merged HubSpot Real estate How a Home Renovation Company Merged 1,600+ Duplicate HubSpot Records Created by a Call-Tracking Integration, with Dedupely Read the story](https://dedupe.ly/case-studies/home-renovation-hubspot-call-tracking-duplicates) - [Customer story 2,158 company matches merged in under 7 weeks HubSpot Financial services How a Fintech Gave Its Sales Team a Pipeline It Can Trust by Clearing 4,000+ Duplicate Matches with Dedupely Read the story](https://dedupe.ly/case-studies/fintech-hubspot-duplicate-deals) ### Clean up your HubSpot Companies and Contacts today [Try for free](https://v2.app.dedupe.ly/signup) [Talk to sales](https://meetings.dedupe.ly/meetings/gaby-salinas/talk-to-sales) Skip to content