Translator Multilingual Survey Response Normalizer

Overview

This flow normalizes multilingual survey feedback for centralized analysis using Microsoft Translator V3. When a response is submitted to a Microsoft Forms survey, the flow detects the language of the open-text answer, translates it to a configurable pivot language (default English), scores sentiment in-flow, and stores the original verbatim, the translation, the detected language, and the sentiment in a Dataverse table. Feedback arrives in many languages and is hard to analyze together; normalizing every verbatim to one language while preserving the original lets analysts work in a single language and report consistently across regions.

Use Case

A CX / marketing team runs global surveys and wants every verbatim translated and analyzed centrally, with the original preserved. Analysts read a single Dataverse table that holds both the source-language text and the English rendering plus a sentiment label, so cross-region reporting needs no manual translation.

Flow Architecture

Environment Variables

Connectors & Connections

Connector	API name	Actions used
Microsoft Forms	shared_microsoftforms	CreateFormWebhook GetFormResponseById
Microsoft Translator	shared_microsofttranslatorv	Translate
Microsoft Dataverse	shared_commondataserviceforapps	CreateRecord
HHTTP (built-in)	http	HTTP - Detect language

Note — All connections are referenced as solution connection references; the flow is portable between environments as long as a connection is mapped at import time.

Customization Guide

Almost every realistic variant of this flow can be implemented by changing environment variable values. A few cases require small edits inside the flow definition — those are called out explicitly below.

Pivot language

Change flowlibs_PivotLanguage to normalize to any language Translator supports (e.g. es, fr).

Sentiment quality

Swap the in-flow keyword heuristic for Azure AI Language Sentiment or Azure OpenAI; keep the detected language as a report dimension.

Multiple open-text questions

Duplicate the compose/translate actions per question id, or loop over an array of question ids.

Key phrases / themes

Add a Translator key-phrase or AI Language key-phrase step on the translated text.

PII

Redact PII before translation if required by policy.

Key Expressions

The flow is intentionally light on Power Fx / WDL gymnastics — the heaviest expressions are the branch-name concatenation and the approval outcome check. They are listed below in the order they appear in the flow.

@coalesce(first(body('Detect_Language'))?['language'], variables('varPivotLanguage'))

@first(body('Translate_Text'))?['TranslatedText']

@sub(length(body('Filter_Positive_Matches')), length(body('Filter_Negative_Matches')))

@body('Get_Response_Details')?[parameters('EnvVar_SurveyOpenTextQuestionId')]

Customize & download