Product Updates

Introducing the transform step: AI-powered data intelligence for your workflows

We’re excited to introduce the Transform Step, a new AI-powered workflow step that lets you clean, reshape, and enrich data directly inside any Intellistack Streamline workflow. With this powerful data transformation, you can turn raw inputs into structured, system-ready outputs using plain language, no code, and no external tools.

The result is faster workflows, cleaner and more usable data, less manual work, and no external tools.

AI-powered data transformation, built in

The Transform Step makes data transformation a first-class part of your workflow. Data within a workflow is often collected from various sources, including freeform text fields, semi-structured inputs, and external systems. However, this raw daw is often not immediately useful in a workflow without additional processing, whether that’s reformatting, classifying, summarizing, or even deriving new values.

Instead of exporting data, building custom scripts, or relying on third-party services outside the platform to make that data usable, you can now handle common and complex transformations exactly where they belong.

Add a Transform Step anywhere data needs to change. Describe what you want to happen in natural language. Map in the input fields and choose where the output should go. The step takes care of the rest automatically for every submission.

Example of using the Transform Step for understanding sentiment on free form text.

Practical use cases you can use right away

The Transform Step was built from real customer needs. Here are a few ways teams are already using it.

Before Transform Step

After

The Impact

Generate professional identifiers automatically

Admissions staff copy-paste and trim email addresses to create usernames, one applicant at a time. Peak season means hours of tedious manual work.

The Transform Step extracts everything before the @ symbol automatically.

500+ applications processed daily. Zero manual data entry.

Standardize messy input

Sales reps enter job titles inconsistently: "ceo," "CEO," "Chief Executive Officer." Reports break. Searches miss contacts. Revenue ops cleans data for hours before every board meeting.

Job titles normalize on submission. Reports and searches work the first time.

Hours reclaimed from data cleanup. Revenue ops focuses on strategy, not fixing spreadsheets.

Route work based on intent and sentiment

Support agents manually read every message to determine urgency and route tickets. Critical issues wait in queue while agents triage.

The Transform Step analyzes sentiment and urgency in real-time, tags tickets, and routes urgent issues to senior support.

Urgent cases reach the right team in minutes, not hours. Faster resolutions, happier customers.

Create structured summaries from field notes

Technicians write: "Arrived at 10, replaced pump, tested system, all good." Operations managers manually parse notes to update dashboards and client reports.

Field notes become structured summaries automatically—ready for dashboards and reports.

Real-time visibility. No manual parsing. Faster client updates.

Extract insights from open-ended feedback

Product teams get paragraphs of feedback but can't read or categorize every response. Feature requests and pain points get buried. Decisions lack complete data.

The Transform Step identifies sentiment, themes, product mentions, and feature requests—turning paragraphs into tagged data.

Product decisions backed by complete customer voice. Trends spotted fast. Features prioritized with confidence.

Generate system-ready formats

Teams manually reformat report titles for filenames, URLs, and API calls—converting "Quarterly Revenue Analysis Q2 2025" into "quarterly-revenue-analysis-q2-2025" character by character.

The Transform Step seamlessly converts human-readable titles into clean slugs automatically.

Zero formatting errors. Files, URLs, and APIs work correctly every time. Integrations that just work.

Example of using the Transform Step to analyze urgency in real-time.

How the data transformation works

The Transform Step fits seamlessly into your workflow between any data source and any downstream action, such as when needing to bring data from one form to another, but in a different format. Each transformation lives in its own step, making workflows easier to understand, test, and maintain. You can even chain multiple Transform Steps together and reference outputs from earlier steps.

You define the transformation using a natural language prompt, test it with sample data, and then let it run automatically on every submission.

  1. Add the step where you need it
  2. Describe what you want in plain language (example: "Extract the username from this email address")
  3. Map your input fields
  4. Test your prompt
  5. Choose where the output goes
  6. Done—it runs automatically on every submission

Start transforming data today

These examples are just the beginning. The Transform Step is flexible by design, so you can adapt it to your own processes, formatting standards, and data needs. If you can describe the transformation, you can build it.

The Transform Step is available today for all Intellistack Streamline users. Add it from the step library and start building smarter, more efficient workflows with AI-powered data intelligence built right in.