Why your HubSpot import is failing

First, get the real error

HubSpot does not fail an import as one lump. It imports the rows it can and lists the rows it could not, with a reason against each. That list is the useful part, and it is easy to miss.

Open the import from the imports list and download the error file. It gives you the failing rows with the error name attached, which turns a vague failure into a specific fix. Everything below is keyed to those error names.

Worth knowing before you go back to the spreadsheet: some of it can be fixed inside the import tool. At the mapping stage you can replace a value and HubSpot applies the change to every row in that column where it appears, which handles a misspelling repeated across hundreds of rows. That works well for one or two bad values. It stops being the quicker route when several columns need work, or when the same file arrives every month.

If the whole import was rejected before it got to the rows, it is almost always one of the file-level problems in the next section. HubSpot checks the shape of the file before it looks at the contents.

Error message lookup

Find the message from your error file below, then go to the section that fixes it. The error names are HubSpot's own, as its Review and troubleshoot record import errors article lists them (updated September 25, 2026). The table covers the errors this page fixes; that article lists every one and says what it means, and this page says how to fix the file.

The file itself was rejected

Most of these stop the import before a single row is read; Row data too large skips only the row that is too big. They are the quickest to fix and the most common cause of a file that will simply not upload.

A file refused for having far more rows than you can see is carrying rows that only look empty: formatting or a stray value far below your data makes them count. Press Ctrl+End to see where Excel thinks the data ends, delete the rows and columns past your data, then save and check again. A file that really is too big has to go in as several; see how to split a large file for a CRM import.

Your columns did not match HubSpot

These are about the header row and the properties behind it. HubSpot has to turn each of your column names into one of its own properties, and it will tell you when it cannot.

Individual values were rejected

These are the ones that let most of the import through and quietly drop the rest, which is why an import can look successful and still leave you short.

Duplicates and identifiers

HubSpot matches on a unique identifier, usually email for contacts and domain for companies. Most of these errors mean your file disagrees with itself about which record is which.

Email matching is case-insensitive in HubSpot, but a leading or trailing space is not the same address. If updates are landing as new contacts, check for whitespace before you check anything else.

Importing companies and contacts together

You can bring contacts and their companies in with one file or two. In one file, each row holds a contact and its company, with an identifier for each: Email or Record ID for the contact, Company domain name or Record ID for the company. Leave the identifiers out and HubSpot creates duplicate records instead of linking each contact to its company.

A contact who belongs to two companies takes two rows, with the same email and a different company domain in each. With two files, one for contacts and one for companies, both need a column in common that connects the records. If both files carry Record IDs, name the columns apart, for example Record ID - Contacts and Record ID - Companies.

Company domain name keeps only the domain: anything after it, such as /about, is removed on import. When one domain matches several companies already in HubSpot, HubSpot leaves that company unchanged and lists Multiple companies with this domain. Merge or delete the extra companies, or import with Record ID instead.

With Association record not found, Invalid association identifier, Failed to create association, Association label not found and Association limits exceeded, the records still import. Only the link or its label is missing, so you can fix the file and import again, or link the records by hand in HubSpot.

It said it worked, but nothing changed

This is not really an error, and it catches people out more than the errors do. HubSpot asks whether the import should create records, update them, or both, and it does exactly what you chose.

Create only import means your file matched existing records and HubSpot left them alone, as instructed. Update only import means the opposite: rows that matched nothing were skipped rather than created. If you meant to do both, choose create and update, and run it again.

The ones you cannot fix in the file

Some errors are about your HubSpot account rather than your spreadsheet, and no amount of cleaning will move them.

Limit exceeded means you have hit your record allowance. GDPR blocklisted email means those contacts cannot be added by import at all, whatever the file says. Those two are HubSpot-side decisions, and the file is not the lever. Association errors are mostly about the file, and they are under Importing companies and contacts together.

The imports that do not fail

Every error above announces itself. The imports that cause real trouble are the ones that succeed. Nothing is rejected, the summary looks healthy, and the damage only surfaces days later when a report is wrong or a rep asks why a contact went cold.

Experienced HubSpot admins consistently name the same handful. An import can overwrite Original Source, which is the field your attribution reporting rests on. It can move Lifecycle Stage, sometimes backwards. It can replace tidy dropdown values with free text typed by whoever built the file. It can break company associations, change record owners and the routing that depends on them, and land values in the wrong property entirely because two columns had similar names.

The one that surprises people most: an import can trigger workflows. If enrollment criteria match the records you just touched, automation fires on all of them at once, and that can mean live emails going out to a few thousand people who were not expecting them.

The question worth asking before any import is not whether the file is clean. It is: what is this file allowed to change? Everything else follows from the answer.

Importing an Apollo export into HubSpot

An Apollo people export usually imports into HubSpot with few errors. The trouble is in what it brings with it: more than 70 columns, some named so that HubSpot maps them to the wrong property. These are the ones to deal with before you upload.

1. Check for the locked placeholder email

Apollo now asks you to access each email, at a credit apiece, before it exports a contact. Older exports can still carry rows where every unrevealed email is the same placeholder, email_not_unlocked@domain.com. HubSpot imports the first of those rows and rejects the rest with a Duplicate alternate ID error, so only one of those people arrives. Filter them out, or reveal the contacts in Apollo and export again.

On the mapping screen, before the import. After it, the same rows are listed as Duplicate alternate ID.

2. Decide what to do with Email Status

Apollo marks each address, for example Verified or Unverified. HubSpot has no field for it and doesn't suggest one. Import only the Verified rows, or create a custom contact property and map the column to it so you can filter later.

3. Choose which phone goes where

Apollo exports five phone columns: Work Direct, Home, Mobile, Corporate and Other. HubSpot maps Mobile Phone to Mobile Phone Number on its own. Map one work number to Phone Number by hand. Corporate is usually the switchboard, so Work Direct is the better choice when it is there.

4. Keep person and company columns apart

The person's location is in City, State and Country, next to the company's Company Address, Company City, # Employees and Annual Revenue. Map the person columns to the contact. The company columns belong on the Company record, in a separate company import.

5. Check every row HubSpot maps for you

HubSpot fills in the mapping from the column names and can get it wrong: in our test it mapped Company City to Company size. Go down the list before you click Next. To skip a column, choose Don't import column in its Import as dropdown, then check Continue without mapping all columns.

6. Set Lifecycle Stage and Lead Status yourself

The export carries neither. Its Stage column is Apollo's own stage, such as Cold, so don't map it to Lifecycle Stage. New contacts get your account's default lifecycle stage and no Lead Status, so add both columns to the file, for example Lead and New, to land them in the views and workflows that filter on them.

7. Match Account Owner to HubSpot users

Account Owner holds the Apollo user's email address, and the export's Contact Owner column holds the same one. If that address belongs to a HubSpot user, map Account Owner to HubSpot's Contact owner property. If not, the mapping screen says Could not find owner and lets you pick a HubSpot user for each value. Left as No owner, the contacts import without one.

Operelio's Apollo.io to HubSpot template makes most of these choices for you: it maps Mobile, the first work number and the person's City, State and Country, leaves out Home Phone and Apollo's Stage, sets Lead and New, and puts Seniority, Departments and Apollo lists into Notes with their labels.

A checklist that prevents most of it

None of this is difficult. It is the difference between an import you can undo and one you cannot.

1. Export a backup of the records you are about to touch

Before anything else. HubSpot's restore tool can roll an import back within 14 days, but only a Super Admin on a paid hub can run it, and property history only tells you what changed one record at a time. An export you already have is worth more than any amount of care afterwards.

2. Run a sample first

Twenty or thirty rows. Import them, then open two or three of those records and read the property history to see exactly which fields the import wrote. This is the step people skip and then wish they had not.

3. Decide what the file is allowed to change

Map only the columns you intend to update. A column that is in the file but should not overwrite anything is safer left unmapped than mapped and hoped about. Treat Original Source, Lifecycle Stage, and owner fields as protected unless changing them is the actual goal.

4. Check what automation will fire

Look at which workflows enroll on the properties you are about to write. If the import should not set anything off, turn those workflows off for the duration, or import into a staging property instead.

5. Put untrusted data in its own field

When a list comes from a conference badge scanner or a bought file, the job title in it is a claim, not a fact. Writing it into a separate imported or submitted field keeps your clean data clean and still gets the new information in.

Fixing it once instead of per error

Most of the list above comes from three things: a header row that does not match, values that do not match a dropdown, and duplicates. HubSpot lets you patch some of these in the import tool one value at a time, which is fine once. It is the wrong shape when the same export arrives every month, because you are redoing the same corrections by hand each time.

Clean Headers standardizes the header row and removes the blank and duplicate names behind Incorrect number of columns and Property definition not found. Find & Replace collapses the spelling variants behind Invalid enumeration option. Deduplicate removes the repeated identifiers behind Duplicate unique property value. Excel to CSV takes one sheet out of a multi-sheet workbook, which is Invalid sheet count. The CRM Formatter maps your columns onto HubSpot's real property names and shows a preview of what will be created and updated before anything is written.

Dates have two fixes and it is worth knowing both. HubSpot asks you to pick a date format during the import, so choosing the one your file already uses often solves it without touching a cell. When the column itself is inconsistent, which is the harder case, Data Health Check flags the mixed formats and its Standardize dates fix converts them to YYYY-MM-DD, reading the column in its own day and month order. A date that could be either, in a column that doesn't settle it, is left as it is for you to check.

Originally published in the Operelio how-tos. Which HubSpot import error do you hit most?