Integrations
SmartCompaire is system-agnostic. To test a payroll system migration, you export pay results from the old and new systems, load them into a project, and compare them employee by employee and pay code by pay code.
SmartCompaire does not connect live to SAP, Workday, Oracle, ADP or any other payroll or time system. There are no scheduled pulls, SFTP drop folders or customer API. Every piece of data comes in through the app as an uploaded spreadsheet or CSV, rows typed on screen, or rows pasted from a spreadsheet.
That is deliberate for migration testing. If a system can export pay results by employee and pay code, SmartCompaire can compare it, whether it is the legacy system you are leaving or the one you are moving to. It also means SmartCompaire never writes anything back to either system. It finds, grades and records differences; fixing them stays with your payroll team.
Each project records a source system and a target system with an ID for each. Those IDs label the template columns, so the two exports are easy to tell apart. The upload does not check which system a file came from, so load each export into the matching source or target step. Each organization can also set up which payroll and time systems are offered per country.
Notes on preparing source and target exports for common migrations. Each page describes file-based comparison, not a live connector.
Load SAP and Workday payroll exports, map wage types to pay components, line up tax authorities and compare parallel-run results employee by employee. File-based, with no live connection.
Read moreParallel test an Oracle E-Business Suite or Oracle HCM Cloud payroll migration by mapping elements to target pay codes and comparing exported pay data.
Read moreValidate a move to or from ADP by comparing exported pay results code by code, employee by employee.
Read moreEvery project, whatever the systems, is set up in the same four steps. Each step accepts a file upload, typed entry or rows pasted from a spreadsheet.
Pair each source employee number with its target employee number. The template has two columns titled with your source and target system IDs. Each number can be mapped only once in either direction, and duplicates are rejected with their row number.
Translate source pay codes into target pay codes, each tagged with a Bucket such as Earnings or Deductions and a Bucket code type such as Gross or Pre-Tax. Tax authority codes are optional. Two source codes mapped to one target code are summed and compared as a total.
Export pay results from the system you are leaving into the 15-column pay data template: payroll area, periods, check date, employee, tax authority, pay code, number, rate, amount, and MTD, QTD and YTD amounts. Only PayCode is required on each row.
Load the same layout from the new system, using the target template, whose columns carry the target system ID. A readiness check then shows how many rows will match a mapping before you start a run.
Learn morePayroll extracts are rarely tidy. The import is built to accept what systems actually produce and to tell you exactly which rows it could not use.
Each setup step offers a downloadable Excel template with the column headers the upload expects. Header matching ignores case, spaces and punctuation. Keep only the template's columns in your file: an extra column whose name ends like a template column, such as "Check Number" or "Gross Amount", can be read in place of the template column.
Upload .xlsx, .xls or .csv files; only the first worksheet is read. CSV files can be UTF-8 or Windows-1252, separated by commas, semicolons, tabs or pipes.
Leading zeros in employee numbers are kept when the column is exported as text, as in CSV. Amounts with $ signs, thousands commas or accounting negatives such as (12.00) are parsed and stored as exact decimals, not floating point.
Check dates are accepted as Excel serial numbers or as YYYY-MM-DD, MM/DD/YYYY, DD/MM/YYYY, YYYY/MM/DD, DD-Mon-YYYY and MM-DD-YYYY.
Every upload is imported in the background with progress tracking and a cancel option, 5,000 rows per batch. Files over 2 MB skip the in-browser preview. The default upload limit is 100 MB.
Each rejected row is reported with its real spreadsheet row number and a plain reason. The first 100 are shown on screen, and up to 100,000 are in a downloadable CSV with the original cells, so you can see exactly what to correct.
A project is either a payroll or a time comparison. Time projects use the same file layout and the same comparison logic, with hours or units in the Number column. Number differences are recorded for every row, but only the Amount difference is graded against thresholds.
Because mappings live in SmartCompaire rather than in either system, a mistake is cheap to correct. Each comparison run re-matches pay rows against the current mappings, so fixing a paycode mapping takes effect on the next run without uploading the pay data again.
A few things are worth knowing before you plan your extracts.
No. SmartCompaire works from files exported from those systems, or from rows typed or pasted in the app. There are no live connections, scheduled pulls, SFTP folders or customer API.
Any system that can export pay results by employee and pay code to Excel or CSV. You arrange the export to match the pay data template, then map employee numbers and pay codes between the two systems.
SmartCompaire handles it in the employee mapping step. You pair each source employee number with its target number, and the comparison joins pay rows through that mapping. Employees who appear in pay data without a mapping are listed, up to 200 per side, with their row counts so you can add them.
Yes, for typical extracts. Files are imported in the background in batches of 5,000 rows, with progress tracking. The default upload limit is 100 MB per file, and the import streams rows in batches rather than loading the whole file, so large extracts do not need to fit in memory.
No. It reads the files you upload and records differences, statuses, defects and comments in SmartCompaire. Nothing is written back to any payroll or time system.
Bring a source and a target extract from your parallel run; we will walk through the comparison with you.
Request a demo