NEW
MCP Alteryx parser
AI
Proof
Pricing
Book a demo
Scan your code free
Hard sources — not covered by free tools
Also parsed — certify what free tools miss
Runtimes
After migration
Pricing
$10,000 a year for the platform. Conversion is metered in MigryX Conversion Units, as you go or prepaid for less. People are a separate line, only if you want them. Your price is set after the Free Assessment.
The same price whether the output is Databricks, Snowflake, dbt, BigQuery, or PySpark.
$10,000 a year
Runs in your cloud or data center. Updates, support, parsers for your sources, basic discovery, and re-runs of what you already converted. No conversion units included. Basic discovery is a file inventory, line counts, a complexity mix, an MCU estimate, and a program list.
$75 per MCU
One MCU is one simple SAS program of about 500 lines. Every other source counts at its own published rate. Billed monthly for what you convert.
Up to 50% off
Buy a pool of MCUs whenever you are ready and pay less per unit. The more you buy, the lower the rate. Anything past the pool is back at list.
Regional pricing. Installations licensed in India, Southeast Asia, Latin America, and Africa pay 35% less on both the platform license and MCUs.
A few quick questions. Rough numbers are fine. The Free Assessment counts your real code and sets the price.
A small tool you run on your own machine. It counts and sizes your code and tells you the MCUs. Nothing is uploaded.
Point it at a folder of code. No database, no server, no install beyond Python. It writes one report you keep, or send to us for a quote.
The parse that shows how the estate fits together. It draws MCUs. Those MCUs are credited back on files you convert within 12 months.
Conversion is a separate MCU charge: the generated code and the row-level parity check. It is not included in advanced analysis.
Buy ahead at any time. The whole pool gets the rate of the tier it reaches. Prepaid MCUs stay valid for 12 months from the day you buy them.
| MCUs bought ahead | Discount | Price per MCU |
|---|---|---|
| As you go | None | $75 |
| 100 or more | 10% | $67.50 |
| 250 or more | 15% | $63.75 |
| 500 or more | 25% | $56.25 |
| 1,000 or more | 30% | $52.50 |
| 2,000 or more | 40% | $45 |
| 5,000 or more | 50% | $37.50 |
Never inside the license or the MCU price. Pick none, offshore, onsite in the US, or both.
You convert, run, reconcile, and deploy into your own Databricks, Snowflake, or dbt project. Support comes with the platform license.
Our engineers in India, with overlap into your working day. Parity runs, remediation, reconciliation. Quoted in hours.
US-based engineers at your site or on your network. For regulated environments and cutover. Quoted in hours.
A separate product with its own subscription. Not metered in MCUs.
Starting at $50,000 a year
Lineage, change intelligence, and evidence across your databases, models, and users after conversion.
A MigryX Conversion Unit. One simple SAS program of about 500 lines is one MCU. A simple Alteryx workflow is less than one; a COBOL program is several. Complexity raises the count. The Free Assessment works it out for your code, so you never convert by hand.
Included with the platform license, and the same scope as the Free Assessment. File inventory and counts by source type. Lines of code. An estimated complexity mix. An MCU estimate and the price. A high-level program list. It does not include column-level lineage, dependency resolution, or exportable documentation.
Column-level lineage and cross-file dependencies. Macro, include, and copybook resolution. A dataset and table catalog with usage mapping. Comparisons of the old system and the new one. Exportable reports and documentation. MCP and AI integration. It draws MCUs from the first unit. Those MCUs are credited back on files you convert within 12 months. This is the definition a statement of work should use.
It comes from rough counts and a guess at complexity. The Free Assessment reads your actual files. The platform’s full parse sets the count that is billed.
No. The same code counts the same MCUs whether it lands in Databricks, Snowflake, dbt, BigQuery, or PySpark.
No. Pay as you go at list, and prepay later if the volume grows. A prepaid pool is cheaper than paying as you go for the same MCUs.
12 months from the day you buy them, whatever your license renewal date. So you can prepay now for work that starts in a few months and still keep your rate.
No. Each client organization gets its own platform license, with the client named on it. Partners who run several client installations draw MCUs from one shared pool, so their prepay rate reflects their total volume with us.
One platform license per client installation, and one MCU pool across all of them. Your prepay rate is set by the MCUs you prepaid across every installation in the last 12 months, including the purchase you are making. Keep prepaying and you keep the rate. Advanced analysis is metered per client installation and credited back on files that installation converts. Atlas is always licensed to the client directly.
The same way. Each country, department, or business unit runs its own installation with a platform license, and all of them draw from one shared MCU pool. Your prepay rate reflects your total volume across the organization over the last 12 months. Talk to us about platform fees across several installations.
No. Installations licensed in India, Southeast Asia, Latin America, and Africa pay 35% less: $6,500 a year for the platform and $48.75 per MCU, with the same prepay discounts on top. The price follows the country where the installation is licensed, not where the buyer or partner is based. Invoices can be issued in local currency at a rate fixed for the contract term.
A managed SaaS option is coming. Talk to us if you need it now.