Smart Columns let you create new columns for data enrichment, transformation, mapping, or cleaning. The limit is your imagination!

They are computed automatically using:

• simple mappings (ex: to map SKUs to specific values)

• formulas (ex: to combine NPS scores with star-ratings)

• or prompts (ex: to associate emotions to verbatims), so you can describe your needs in plain language.

Define a transformation once and Caplena applies it across thousands of rows in minutes. As new data flows in, Smart Columns update automatically.

Turn messy, inconsistent data into clean categories.

Example: Turn US, USA, United States, and U.S.A. into one consistent value.

Map country codes to full names, normalize product variants, or group free-text labels into unified segments. Just define the mapping (or let the AI suggest one) and watch your data fall into place.
Instantly, across every row.

Concatenate fields, perform calculations, apply conditional logic, or reformat dates — right inside Caplena.

Example: Combine region and product line into a single segment key for cross-tabulation.

Formulas use a simple template language that references any column in your project.
No code, no export, no hassle.

Not sure how to write a formula or build a mapping? Just describe what you want in plain language. Caplena's AI generates the configuration for you : ready to preview, fine-tune, and apply.

A few examples of what you could to:

• Entity extraction: extract product names from text comments

• Profanity filtering: remove slurs, offensive or inappropriate language

• Segmentation: cluster audiences based on verbatim (ex: occasional vs. recurring buyer)

• Emotion analysis: assign an emotion based on open-ended text

Go from idea to enriched data in seconds.

Smart Columns significantly reduce the effort required to prepare your data.
Like a built-in data enrichment tool, they help you clean, enrich, and transform your data.

Even without data engineering skills, you can augment your dataset for deeper insights.

The easiest way to bring your data into Caplena is through native integrations with the feedback collection tools you already use (like Qualtrics, Medallia, or Survey Monkey). Teams can connect their sources and start analysing feedback in minutes, without writing a single line of code.

For teams that want to automate data flows or build Caplena into a custom stack, the Caplena REST API lets you push data in from any source programmatically (in real time for live use cases, or in bulk for offline workflows).

And if you need to get started quickly, or to analyze an ad hoc research project, you can always upload files directly in CSV or SPSS format.

However you bring your data in, it all lands in one unified workspace: ready to transform, enrich, and analyse.

Yes, you can export topic and sentiment assignments, as well as your columns - including Smart Columns. A nightly BigQuery export let's you easily transfer results in bulk to your data warehouse and analyze them in your favorite BI tool such as PowerBI & Tableau.

Yes! Once your data is imported into your Caplena project, any source column can be referenced to create Smart Columns. This means you can achieve custom combinations & normalize your data at scale.

If you're comfortable with Excel formulas, you'll feel right at home. Formula mode works on the same principle, but uses a syntax called Jinja - a format that reads close to plain English and is widely used across analytics tools.

The real upside is flexibility: Jinja handles numbers, text, dates, and conditional logic all in one place. If you can write an IF statement in Excel, you can write a Jinja formula in Caplena - no IT involvement needed.

Caplena lets you preview results on a sample of your data before applying them at scale. That way, you can refine your instructions until the output looks exactly right.

For structured outputs (like category labels or sentiment scores), you can define allowed values to keep results consistent across thousands of rows.

Smart Columns run alongside your existing data. Your original data or text-to-analyze responses are never altered or overwritten. You can review, adjust, and re-run any enrichment column at any time, making it easy to iterate without any risk to your source data.

We're taking care of all the gruesome reliability, load and performance issues, you only have to deal with the semantic part.

No. Your data is not used to train Caplena's AI or any third-party model. For AI features, we use frontier models from Google Cloud and Microsoft Azure, running in their EU data centers. Caplena may use usage data (how the product is used) to improve our services, benchmark performance and test new AI algorithms. See our Trust Center for all subprocessors and where they process data.