How 7 operators use custom data transforms to automate complex processes

When asked about Parabola’s newest AI feature, Janie and Jack’s Director of Transportation & Operations, Samantha Mandel, answered in four simple words: 

“I LOVE THIS STEP.” 

Based on dozens of submissions from other customers, Samantha’s not alone.

We launched the Custom Transform Contest to celebrate Parabola’s new Custom transform step—an AI feature that allows you to turn plain language instructions into custom data transformation steps. In the words of our CEO, “it’s honestly magic.” 

The rules of the contest were simple—submissions were evaluated based on how creatively users could solve real, complex data challenges using the new step. 

For the winner? $250 in cold, hard cash. 

For the runners up? Brand new Parabola swag. 

…and bragging rights for all. 

Through these submissions, we learned about operators cutting out 50+ steps from their workflows, creating net-new visibility into ERP data, and improving SLA compliance. With so many incredible submissions, we struggled to pick just two—so without further ado, meet your winners: 

‍First place

• Chad Coley: Inventory Manager at Harry’s
  
  • Use case: ERP-to-DC reconciliation

Runners up

• Dominic Seminara: Ecommerce Manager at Cold Cuts Prints
  
  • Use case: Shipping zone & weight analysis
• Brian Zhang: Strategic Finance at Anine Bing
  
  • Use case: Forecast vs. actual sales summary
• Deven Hidalgo: Manager, Operations Strategy at Maiden Home
  
  • Use case: SKU extraction and lead time calculation
• Austin Schebaum: Associate Program Manager at Uber Freight
  
  • Use case: SLA performance evaluation
• Samantha Shackett Director of Transportation & Operations at Janie and Jack
  
  • Use case: Excel cleanup & enrichment
• Richard Tran: Business Intelligence Manager at Daylight Transport
  
  • Use case: Fuel transaction timestamp normalization

Before state 

Before we get into submission details, let’s talk about how operators were solving these problems in the pre-Custom-transform era. 

The story is almost identical for every respondent—before this step, their use case was possible in Parabola, but possible ≠ simple and straightforward. Operators were using anywhere from 2 to 50 additional steps to perform the same data transformation that they’re now achieving in one step—making Flows more complex to understand, more difficult to build, and harder to maintain over time. 

Let’s see what the new world looks like with the Custom transform step. 

Meet the winner

Chad Coley: Inventory Manager at Harry’s

About the brand

Harry’s is a men’s grooming brand that pioneered the D2C model—offering high quality shave, body, hair, and skin care products for reasonable prices. 

Use case

ERP-to-DC reconciliation

The challenge

Chad needed to reconcile transactions between his company’s ERP system and the distribution center’s transaction report—highlighting any discrepancies across shipments, receipts, or adjustments. Making the task even more difficult, Chad’s data wasn’t standardized across sources, so he needed to join and clean the data before performing the reconciliation. 

The solution

Using the Custom transform step, Chad consolidated logic that previously spanned many steps into a single prompt. The step creates new columns, sets new column values based on complex conditional logic, and removes data they no longer need. In Chad’s words, “This approach effectively created an XLOOKUP-like function within the ‘Customer’ column by referencing previous entries for the same PO number. It preserved non-blank entries and, for blank entries, filled in the MAX value for that PO in the column.” 

[This Flow] has been invaluable for investigating variances…I’ve found this new step to be extremely helpful in reducing the number of steps required to solve more complex data problems. It’s great at removing the friction of overly complicated logic and gets to the outcome faster.”

Business value

This step allows Harry’s to uncover variances in their ERP faster and more consistently, improving the data quality in their ERP and ultimately supporting SLA compliance. Additionally, by reducing the complexity in his Flow, Chad made variance reporting workflow easier for the broader team to manage and improve over time. 

Runners up

Brian Zhang: Strategic Finance at Anine Bing

About Anine Bing

Anine Bing is a Los Angeles–based fashion brand focused on blending elements of Scandinavian style with American fashion, found in 90+ countries around the world. 

Use case

Forecast vs. actual sales summary

The challenge

Brian needed to consolidate actual historical sales data and reformat it to align with their forecasting template. This required grouping data by quarter and year, calculating key metrics, and matching the structure used by the finance and planning teams.

The solution

With the Custom transform step, Brian generated a dynamic summary following his team’s internal template. The step grouped sales data by quarter and year, aggregated relevant metrics, and output a dataset formatted to match the forecasting model.

Business value

After proving out this forecasting approach in Parabola with one step, Brian can now roll out similar forecasts to other parts of the business to support planning activities. 

Deven Hidalgo: Manager, Operations Strategy at Maiden Home

About Maiden Home

Maiden Home is a modern luxury furniture brand that delivers handcrafted, made-to-order pieces directly to customers.

Use case

SKU extraction and lead time calculation

The challenge

Deven was facing a classic operator challenge: She needed to join unstandardized SKU data from multiple sources so that she could calculate lead times and communicate results to her business partners. This involved tons of complex conditional logic and data extraction rules to clean and join her datasets. 

The solution

By using 14 Custom transform steps in her Flow, Deven was able to cut the complexity from her old Flow and replace it with atomic steps that simply perform each of these functions, simply described in plain language. 

I love that this feature saved me so many steps and let me re-use logic in a much cleaner way.”

Business value

At the end of this Flow, Deven creates a master lookup table that’s leveraged by the business at large. Using this data, Deven and Maiden’s business partners can stay on top of real-time cost, lead time, and inventory data. 

Austin Schebaum: Associate Program Manager at Uber Freight

About Uber Freight

Uber Freight empowers shippers with a comprehensive suite of logistics solutions, combining advanced technology with an extensive carrier network to optimize every step of the freight lifecycle. 

Use case

SLA performance evaluation

The challenge

While comparing reported appointment dates against the latest required delivery date, Austin needed to perform a date comparison and reformat data values in his Flow. By executing this comparison, he could proactively alert the appointment scheduling team in real time about any shipments that may require manual intervention.  

The step has been great! Enjoyed seeing what it is capable of.”

The solution

Using the Custom transform step, Austin wrote a single prompt to compare delivery and expected dates, apply threshold logic, and flag records for human intervention. The step condensed everything into one place, simplifying the build and making the SLA logic reusable across other Flows.

Business value

Uber Freight is already an automation powerhouse. With this step, they’re adding to their collection of re-usable elements, making it easier to continue scaling up their automations across the business. 

Samantha Schackett: Director of Transportation & Operations at Janie and Jack

About Janie and Jack

Janie and Jack is a modern children’s apparel brand offering elevated, timeless fashion for kids.

Use case

Multi-column Excel cleanup & enrichment

The challenge

After converting a partner’s .txt file to an Excel file, Samantha needed to clean and enrich a dataset full of blank values. These cells needed to be filled based on prior row values and required transformations across three separate columns—including updating a date, changing a value, and inserting a symbol—all based on conditional logic.

Previous approach

Coming from the world of Excel, Samantha knew the exact function to use for this transformation in Excel—but as a brand new Parabola user, she was still learning about the best step for each transformation. 

I LOVE THIS STEP. Will definitely be using this in the future!”

The solution

Samantha was able to creatively leverage her knowledge of Excel to build an effective solution in Parabola. In her Custom transform step, Samantha provided the formula that she knew would work in Excel: =IF(A2="",A1,A2)

Since Parabola’s step can handle just about any type of instructions, the step knew exactly what Samantha was trying to accomplish based on her Excel description—which perfectly filled in her columns. 

Business value

This allows Samantha’s team to handle messy vendor files and transform data from external systems more efficiently. Ultimately, this enables faster ingestion, reduced manual error, and streamlined reporting handoffs to downstream stakeholders in transportation and logistics.

Richard Tran: Business Intelligence Manager at Daylight Transport

About Daylight Transport

Daylight Transport is a leading expedited less-than-truckload (LTL) carrier in the United States, known for its high-speed, reliable freight services.

Use case

Fuel transaction timestamp normalization

The challenge

Richard needed to process fuel transaction data from a vendor who delivered the date and time in two separate columns. To prepare the data for ingestion into a third-party platform (Samsara), he had to combine the date and time, convert it to military time format, and apply a time zone adjustment.

The solution

With Custom transform, Richard condensed all 16 steps into a single prompt. The transformation logic merged and reformatted the data in one go, drastically simplifying the Flow and making it far easier to modify and replicate across similar data pipelines.

I will definitely implement this new step in a lot more flows!”

Business value

This transformation simplifies Flow maintenance and ensures timestamp formatting is standardized before pushing data to critical external systems like Samsara—boosting data accuracy and reducing friction in the integration.

Dominic Seminara: Ecommerce Manager at Cold Cuts Prints

About Cold Cuts Prints

Cold Cuts Prints is a fast-growing ecommerce brand that specializes in high-quality custom art prints. 

Use case

Shipping zone & weight analysis

The challenge

Dominic was tasked with analyzing which U.S. states and countries Cold Cuts Prints ships to most frequently, along with the average weight per shipment. Using this data, he could help ensure the team was always obtaining competitive shipping costs, which is much easier said than done when working with messy carrier data. 

This was possible to do without Custom transform, but it would’ve taken 50+ steps. Now it’s fairly simple—and impressive.”

The solution

With a single Custom transform step, Dominic can now group shipments by state and country; calculate average weights; and get clean, actionable insights as an output. What once required 50+ steps was now condensed into one, making the Flow dramatically simpler and faster to iterate on.

Business value

This transformation enables Cold Cuts Prints to continuously monitor their shipping patterns to improve cost efficiency and, in their words, discover new market opportunities. It empowers the team to make strategic decisions around fulfillment and regional marketing, using operational data in smarter ways.

—

To learn more about additional use cases for Parabola’s Custom transform step and explore prompting best practices, check out our launch announcement doc. 

Congrats to our winners and happy building!