Avoiding duplicates when importing companies into HubSpot

[TChazot]

Hi everyone,

I want to update a large list of companies from an Excel file into HubSpot. Many of them already exist in our portal, but the names don’t always match exactly. For instance:

• Excel: “XXXX, PLLC”

• HubSpot: “XXXX, PLC”

Because of these small variations, I’m the import will create duplicate company records instead of updating the ones we already have.

Could you advise on the best way to:

1. Match the incoming companies to existing records despite minor name differences

2. Update those records with the new data

3. Prevent any duplicates from being created

Any tips or resources would be a huge help!

Thanks a lot

[karstenkoehler · Accepted solution]

Hi @TChazot,

HubSpot deduplicates companies based on the domain name or record ID: https://knowledge.hubspot.com/records/deduplication-of-records#deduplicate-companies-by-company-doma...

In other words, if your existing records HubSpot have the company domain name maintained and you're also mapping and importing it, HubSpot should not create any duplicates.

The company name itself is not considered at all. Imports without domain or record ID will always create duplicates (unless you have set up other identifiers).

Do you have the company domain name?

Without it, you'll have to first export all companies so you can match the existing company record IDs to the companies in your new import file by company name. You can then map the record ID upon import to avoid duplicates. Different spellings will however create duplicates, that you'll have to look into manually unfortunately.

Best regards!

[Nhal]

The best way to update existing companies in HubSpot without creating duplicates is to rely on unique identifiers like the company domain name or HubSpot record ID rather than the company name, since HubSpot's deduplication logic does not account for slight name variations. Before importing your Excel file, try to enrich your spreadsheet with domain names or export existing company data from HubSpot to match and map record IDs manually. If domains aren't available, use the HubSpot record ID as your unique key for updating. You can also preprocess your data in Excel or another tool to standardize naming conventions before import, though this still won’t guarantee deduplication unless a unique identifier is used. Unfortunately, HubSpot won't detect that 'XXXX, PLLC' and 'XXXX, PLC' are the same, so without domains or IDs, you risk creating duplicates. If this becomes a recurring issue, you might consider using external tools or scripts for fuzzy matching before importing to ensure you're aligning records accurately.

[JRivera78]

Hi,

Our company was generating a lot of duplicates due to the import process. I always select ''Create and update contacts- Create and update Companies.'' This way, I avoid creating duplicates. Remember to have your mapping correct before importing your files. I use Company Name, Website URL (for the company's website), which seems to help me reduce the amount of duplicates. 

I hope this helps, I'm also learning! 

[Rajib_alimul]

Well, this is always a challenge. I have been doing this for so many years, and it seems like it is always different for each case! But I always prefer the import option "update existing companies only" so that  I do not end up creating new ones!

Have you already tried using domain/website address field and HubSpot company ID already?

[victoriahcw]

Great question! To avoid creating duplicates when importing companies, the best approach is to use a unique identifier like the company domain or HubSpot Company ID instead of relying on the company name, which can vary slightly (as you’ve seen). Before importing, try to match and enrich your Excel file with the correct domain or ID from HubSpot—this ensures updates are applied to existing records rather than creating new ones.

If you’re managing large datasets, Coefficient can help streamline this process. It connects Excel or Google Sheets directly to HubSpot, so you can pull existing company records (including IDs), match them to your new data, and update HubSpot without risky manual imports. It’s a great way to ensure data accuracy and avoid duplicates at scale. Coefficient also has a AI Sheets Assistant that may be able to support with identifying patterns and similarly-named domains if you are working with Google Sheets. 

[Josh · Accepted solution]

Hi @TChazot,

If you have domain names, you can use that to ensure they don't duplicate. Or, if you have any other unique identifier, such as a customer number, that can or does exist in both places, that can also be used to de-duplicate. 

If neither of the above are options, there may be some manual work required. You could download the company list from HubSpot and include the company record ID as a column for your new import where those same records exist. In that case they would just update existing records on import vs creating new ones.

If we're looking at a lot of records, you may be able to use ChatGPT to help identify duplicates. If you aren't using a private instance, just be careful on the type of data that you're uploading. 

Here's the knowledge base article on the import tool: https://knowledge.hubspot.com/import-and-export/understand-the-import-tool

That goes into other articles about using record ids or unique values as well.

Josh

[karstenkoehler · Accepted solution]

Hi @TChazot,

HubSpot deduplicates companies based on the domain name or record ID: https://knowledge.hubspot.com/records/deduplication-of-records#deduplicate-companies-by-company-doma...

In other words, if your existing records HubSpot have the company domain name maintained and you're also mapping and importing it, HubSpot should not create any duplicates.

The company name itself is not considered at all. Imports without domain or record ID will always create duplicates (unless you have set up other identifiers).

Do you have the company domain name?

Without it, you'll have to first export all companies so you can match the existing company record IDs to the companies in your new import file by company name. You can then map the record ID upon import to avoid duplicates. Different spellings will however create duplicates, that you'll have to look into manually unfortunately.

Best regards!