Flatfile vs Dromo vs OneSchema

Flatfile, Dromo, and OneSchema address the awkward step between receiving a customer's file and accepting its rows into your system. Flatfile's site now says the company has renamed itself Obvious.✓ Its documented flow centers on matching file headers, then reviewing and correcting data.✓ Dromo describes an embedded importer that highlights cell errors before returning schema-matched JSON to the host application.✓ OneSchema starts with reusable templates that specify the output shape and validation rules.✓ Choose by testing the failure cases your CRM receives, not by counting feature labels. 1235

• Start with Flatfile when mapping transformation is the hard part. Its mapping page describes combining source fields, splitting a source field, and saving mapping decisions.✓ 2
• Start with Dromo when the correction experience inside your product matters most. Its page describes an inline grid, an error count by column, and an application callback after review.✓ 3
• Start with OneSchema when reusable validation templates are central. Its documentation describes templates used across Importers and FileFeeds, required fields, unique values, and row constraints.✓ 5

3 import tools: where the handoff differs

These products should be compared on one controlled CRM import, not on three polished vendor examples. Give each a file with a title row, a renamed column, a blank required field, a repeated identifier, and a value that must be checked against your application. Record what the user can fix, what the administrator must configure, and what your application receives. The vendor pages establish candidate functions for that trial; they do not establish which product will finish it fastest.

Flatfile's overview identifies adding data, matching headers, and validation or correction as the major stages.✓ Dromo's page explicitly walks from upload to header selection, mapping, grid correction, and a result callback.✓ OneSchema's documentation begins with an output template, its rules, an embedded importer, and handling the cleaned data.✓ Those descriptions suggest different setup conversations: transformation rules for Flatfile, in-product correction for Dromo, and reusable schema policy for OneSchema. They are editorial interpretations of the documented workflows, not evidence that any vendor lacks the other functions. 1345

Flatfile: inspect the mapping before accepting rows

Flatfile's help page says a multi-sheet file prompts the user to select the sheet to import.✓ It describes a mapping step that aligns incoming columns with destination fields.✓ More consequentially, it says incoming fields left unmapped are excluded from the import and disappear from later steps.✓ For a CRM migration file, that makes an unmapped custom field a concrete acceptance risk. Build a trial file with a field your team must retain, intentionally leave it unmapped, and check whether the reviewer notices before continuing. 1

The mapping page describes merging several source fields into one destination field and splitting one source field into several destinations.✓ That is useful to test when different account exports put a name or address into different shapes. A trial should include both transformations and a downstream inspection of the actual destination fields. The page also says mapping decisions are saved during work.✓ Test whether a reviewer can stop and resume the same troublesome file without repeating the mapping, then check what happens when the next file uses a different header. 2

After mapping, Flatfile says users can review, validate, and correct values before transformation completes.✓ Create one invalid value and a repeated formatting problem to see how each appears to the person doing the import. The documentation confirms a correction stage; your trial must establish whether the messages explain your specific CRM rule clearly enough to fix the row. 1

Where Flatfile needs a closer check

Trace each source column from upload to the output record, especially columns that are not mapped.✓ Ask who owns the transformation rule, who can change it, and how a revised rule affects an interrupted import. The saved-decision statement is a reason to test that workflow, not a guarantee of how your own governance will work.✓ 12

Migration difficulty for Flatfile

Start with one representative export and a written destination schema. Include a multi-sheet file if your customers supply them.✓ List fields that combine or split, then run the import and reconcile the resulting fields with the source.✓ The documented steps provide a test plan; they do not provide a universal setup duration. 12

Dromo: test the user's error-resolution path

Dromo lists CSV, TSV, TXT, XLS, XLSX, and XML as upload formats.✓ Its page says the user can identify the header row when an export begins with a title or other material.✓ Test both cases with the same data. A file that parses cleanly but treats the wrong row as headers can make every later mapping look wrong, so the first user decision deserves attention. 3

Dromo describes a spreadsheet-style review grid with problems highlighted beside the cells and an error count on each column.✓ Give a nontechnical user a deliberately messy CRM file and observe whether they can find the first blocking issue, understand it, and correct it. The vendor's description supports the controls available; it does not tell you how usable your own error messages will be. 3

Its listed validators include required values, unique values, combinations of fields, regular expressions, length, and conditional requirements.✓ It also describes asynchronous column hooks for checking against your API.✓ Use a duplicate external ID and an account reference that does not exist in your CRM to test local and application-backed rules separately. The page says the `onResults` callback returns JSON matching the configured schema and metadata after the user finishes.✓ Inspect that payload before you design the final write to your CRM. 3

Dromo describes AI column suggestions and optional mapping memory.✓ Its default Private Mode parses and validates the file in the user's browser and sends results to the host application.✓ Its page separately says AI mapping and mapping memory require a mode that shares column names and select values with Dromo.✓ Make the data-handling mode an explicit product decision. Do not assume the AI suggestion and browser-only processing descriptions apply to the same configuration. 3

Where Dromo needs a closer check

Ask the vendor to show the exact mode needed for the mapping behavior you want.✓ Then test an API-backed rule while a user changes a value in the grid.✓ Check what your application receives through the callback and how it handles a row that passes import validation but fails a later CRM write.✓ 3

Migration difficulty for Dromo

Define the expected JSON fields and validation rules, embed one importer, and connect the result callback to a test destination.✓ Run a realistic file through header selection and error correction.✓ Estimate the work from your own rules and integration path rather than from the vendor's feature list. 3

OneSchema: make the output template the contract

OneSchema describes its Importer as an embeddable CSV importer.✓ Its templates define the output schema and the validation rules for clean data, and the same templates can be reused across Importers and FileFeeds.✓ If several CRM onboarding flows should accept the same record shape, test whether one template can govern them without making exceptions hard to manage. Reuse is documented; how many templates your organization needs is a design choice. 45

The template documentation distinguishes a required nonempty cell from a requirement that the user upload and map the column.✓ That distinction matters when an omitted CRM owner field should block the import even if every supplied cell is valid. OneSchema also documents unique values within a column and row constraints based on another column's value.✓ Use a duplicated external ID and a conditional field to see what the user sees and whether the error points to the right row. 5

OneSchema's native validations include number and email syntax checks.✓ For a rule that depends on your own records, its validation webhook can call an endpoint you host and receive errors or warnings.✓ Test the webhook with a known missing account and a known valid one, then inspect how each response is shown. These documents describe rule types and a request path; they do not certify the correctness of your reference data or the behavior of your endpoint. 56

Where OneSchema needs a closer check

Ask an administrator to modify a template, then run the same file through every importer that uses it.✓ Check who can approve a changed required field and whether your receiving CRM endpoint expects the revised shape. For webhook validation, define how your own service returns useful errors and warnings before testing the user interface.✓ 5

Migration difficulty for OneSchema

Write one output template, configure its required and unique fields, embed the importer, and connect the cleaned data to a test destination.✓ Add one conditional rule and one webhook check only after the basic file path works.✓ This sequence is a trial plan, not a quoted implementation estimate. 45

How to decide with one CRM file

Use the same file, destination schema, and reviewer for all three trials. First check header selection and field mapping. Then insert an unmapped custom field, a repeated identifier, an invalid value, and a reference that requires a lookup against your own system. Flatfile's documentation makes the unmapped-field case especially important to inspect.✓ Dromo documents local validators, API hooks, and an output callback.✓ OneSchema documents template constraints and a validation webhook.✓ 135

Record each correction the reviewer makes and inspect the final payload before any production CRM write. A vendor page can tell you that an error grid or validation rule exists; it cannot prove that the rule matches your account schema. Favor the workflow that lets your team explain every rejected row and account for every accepted field. That conclusion should come from your controlled trial, not from an unsourced performance ranking.

FAQ

How we researched this comparison

We reviewed six original vendor documentation and product pages on October 12, 2026: two for Flatfile, one for Dromo, and three for OneSchema. The comparison addresses structured file import into a CRM workflow, not ongoing system synchronization or record merging. Vendor pages support the listed capabilities; the proposed trial steps are editorial guidance for checking fit in a particular implementation.