Blog Home

What Is a CSV Importer? The 3 Types and How to Choose

Albert Aznavour on November 16, 2026 • 8 min read
featured

Takeaways

  • A CSV importer loads the rows of a CSV or spreadsheet file into an application or database. A full importer runs five steps: upload, parse, map, validate, and fix and deliver. There are three types. Product importers and plugins, such as those for WordPress, Drupal, and Jira, load files into one product. Open source libraries such as Papa Parse handle parsing and leave mapping, validation, and error fixing to you. Embedded importers such as Dromo give SaaS teams the whole flow inside their own app. Dromo publishes its price: $599 a month for Professional, with 250 imports included.
A CSV importer is software that loads the rows of a CSV or spreadsheet file into an application or database. A full importer parses the file, matches its columns to the fields the destination expects, checks each value, and lets a person fix errors before saving. There are three types: importers built into a specific product, open source libraries, and embedded importers that SaaS teams add to their apps.

The name covers very different tools. Search for it and the results include a WordPress plugin, a Drupal module, Jira's bulk import feature, a Blender add-on, and products built for software teams. They share a name, and little else. This guide explains what a CSV importer does, how the three types differ, and how to tell which one fits your situation.

What is a CSV importer, and what does it do?

A CSV importer gets the rows of a file into another system. A full CSV importer moves the file through five steps. Simpler tools stop partway.

  1. Upload. A person selects a file or drops it onto the page. Some importers also accept Excel and TSV files.
  2. Parse. The importer reads the raw text and splits it into rows and columns. This is harder than it sounds, because CSV has never had a single agreed format.
  3. Map. Each column in the file is matched to a field in the destination, so that a column headed Email Address lands in the email field.
  4. Validate. Every value is checked against rules: required fields, data types, formats, and uniqueness.
  5. Fix and deliver. Errors are shown so that someone can correct them, and the clean rows are written to the destination.

What changes from one tool to the next is how many of these steps it covers, and how much you have to build yourself.

What are the three types of CSV importer?

CSV importers fall into three groups, built for three different people.

Type and examplesWho it is forWhat you still build
Product importer or plugin
WordPress CSV Importer plugin, Drupal CSV Importer module, Jira's CSV importer
People loading a spreadsheet into one specific productNothing. You format your file the way that product expects
Open source library
Papa Parse, Python's csv module, react-csv-importer
Developers building their own import flowWhatever the library leaves out, such as mapping, validation, and error fixing
Embedded importer
Dromo, CSVBox, OneSchema, Flatfile, Ingestro
SaaS teams whose customers upload dataYour fields and rules, and the code that receives the results

Product importers and plugins

A product importer belongs to one product and loads data into that product only. The WordPress CSV Importer plugin describes its job as importing posts from CSV files into a WordPress blog. Drupal's CSV Importer module imports CSV data into Drupal and lets administrators map CSV columns to entity fields. Jira Service Management has a CSV importer for creating and updating work items in bulk.

If you are trying to get a spreadsheet into a tool you already use, this is the type you want. The trade is control: you get the fields and rules that product supports, and your file has to match its template.

Open source libraries and parsers

Open source libraries are for developers who want to build the import flow themselves. Most of them are parsers: they turn the text of a file into rows and stop there. Papa Parse is a JavaScript parser that runs in the browser and supports streaming, worker threads, and automatic delimiter detection. Python ships a csv module in its standard library that reads and writes tabular data in CSV format.

A few libraries go further. react-csv-importer combines an uploader, a parser, a file preview, and a drag-and-drop screen for mapping columns. Its GitHub repository was archived by its owner on January 5, 2024 and is now read-only, which is the risk with any free component: someone still has to keep it working.

A parser gives you step two of the five. Mapping, validation, and error correction are still yours to design, build, and maintain. Dromo compares the main options in its guide to open source CSV importers, and there is an overview of JavaScript CSV parsing libraries if you want to see what the parsing step involves.

Embedded CSV importers

An embedded importer is a finished import flow that a software company adds to its own product, so that its customers can upload their own data. The customer sees an upload screen, a column matching step, and a review grid where errors are highlighted and can be fixed in place. The software team defines its fields and rules once and receives clean, validated rows.

This type exists because customer files are unpredictable. Your team controls your database, but not the spreadsheet a new customer exports from the system they are leaving. Dromo is an embedded importer. So are CSVBox, OneSchema, Flatfile, and Ingestro. Dromo compares the field in the best CSV importers for SaaS applications, and explains the type in more depth in what an embeddable CSV importer is.

Which type of CSV importer do you need?

The answer depends on whose file it is and where it is going.

  • You want to load a spreadsheet into WordPress, Jira, a CRM, or another tool you use. Use that tool's own importer or a plugin made for it. There is nothing to build.
  • You are a developer loading files you control, such as an internal script or a one-time migration. A parser library is enough. Your files are consistent, so you can skip the mapping screen and write the validation in code.
  • Your customers upload their own files into your product. You need the whole flow, including mapping and error fixing that people who are not developers can finish alone. You can build it on open source parts or embed a product.
  • Files arrive with nobody present, such as a nightly export from another system. That is automated import, which is a different problem from an upload screen. Dromo handles it through its Headless API, a paid add-on to the Enterprise plan.

Why do CSV imports fail?

CSV files fail in a handful of repeatable ways. A good importer catches each one before bad data is saved.

  • There is no single CSV format. RFC 4180, the October 2005 memo that describes the format, notes that no formal specification exists, and Python's csv documentation warns that the lack of a standard produces subtle differences between applications. Delimiters, quoting, and line endings vary with the program that exported the file.
  • Headers do not match. One customer's file says Email Address and the next says Contact Email. Both have to land in the same field, which is the job of the mapping step.
  • Values are in the wrong format. Dates, phone numbers, and currency amounts are written differently from one file to the next. Validation rules catch them, and a review screen lets the person who knows the data correct them.
  • The file is too large for the tool. An Excel worksheet holds 1,048,576 rows by 16,384 columns, so a file with more rows than that cannot be fully loaded into one. Browsers and servers have ceilings of their own, covered in CSV file size limits.

For the most common causes and the fix for each, see why a CSV file will not import.

Should you build a CSV importer or buy one?

This question only applies to the third case, where your customers upload files into your product.

Building makes sense when imports are rare, when the files come from a source you control, or when importing is so central to your product that you want to own every detail. Be clear about the scope first. A parser covers one of the five steps. The rest is a column matching screen, a rule engine for validation, an editable grid for fixing errors, handling for large files, and fixes for every format a customer sends that you did not plan for.

Buying makes sense when customers import often, their files vary, and your engineers have other work to do. Look for a published price. Dromo's Professional plan is $599 a month with 250 imports included and $3 for each additional import. For a fuller comparison of the two paths, see the true cost of building a CSV importer.

What happens when a CSV importer vendor pivots?

An embedded importer sits in your onboarding flow for years, so where the vendor is heading matters as much as what the product does today. Flatfile renamed its company Obvious, after its newest product, an AI agent, and its support site says the importer is unchanged and in active, long-term support. OneSchema's homepage now leads with AI agents for data operations, and its pricing page publishes no price for the importer. Neither company has shut its importer down, but neither leads with it.

Nisto ran Flatfile's importer before moving to Dromo. Its Chief Technology Officer described its experience as a Flatfile customer:

"There's a reason why their docs aren't up to date. There's a reason why their architecture feels old and hasn't changed. It's because they're distracted. They're off doing something else."

Craig Schlegelmilch, Chief Technology Officer, Nisto

Before you choose an embedded importer, ask three things.

  1. Is the importer the first thing on the vendor's homepage, or one item in a list of products?
  2. When were the importer's documentation and changelog last updated?
  3. Is the price published, and what happens to it at renewal?

Importing data files is the only thing Dromo sells. Competitor details in this section reflect each company's own website as checked in November 2026.

How does Dromo work as an embedded CSV importer?

Dromo is an embedded importer for CSV, TSV, XLS, and XLSX files, with SDKs for React, Angular, and JavaScript. A user uploads a file inside your product. Dromo matches their columns to your fields with AI column matching, validates every value against your rules, and shows errors in a grid where the user can fix them. Your application receives the clean rows.

  • Privacy. Private Mode is the default. The file is parsed and validated in the user's browser and the results go straight to your application, so Dromo does not receive the data. AI column matching needs the column names, so it runs in a mode that shares column headers and the distinct values of select fields, while the rows stay in the browser.
  • Validation. Field types cover email, phone number, date, country, URL, and more. Validators add rules such as required, unique, regular expression match, and length, and hooks can check values against your own API.
  • White label. The importer carries no Dromo logo or name on any plan, and its colors and text can be changed to match your product.
  • Compliance. Dromo has completed a SOC 2 Type II audit. HIPAA Business Associate Agreements are available on the Enterprise plan.
  • Limits and price. Professional is $599 a month with 250 imports included, then $3 per import, for files of up to 100,000 rows. Enterprise raises the embedded importer to 1 million rows per file.

Dromo is not the right tool if you only need to load a file into WordPress or Jira, or if a script and a parser already do the job. It is built for the case where your customers bring files you have never seen and need to get them into your product without help. To see it with a sample file, try the live demo.