Maptionnaire has three separate AI-assisted tools for making sense of open-ended, qualitative survey data. This article explains what each one does, how they relate, and which to reach for depending on what you're trying to find out.
The three tools at a glance
- Sentiment Analysis - a quick read on tone. Classifies every open-text response as positive, neutral, or negative and shows the distribution as a chart.
- AI Tagging - structured thematic coding across all your open-text responses, with a written narrative built from the tags. The closest thing to what a human analyst would produce doing this by hand, just much faster.
- AI-powered summarization - a conversational tool for asking specific questions of your data: "what are people saying about parking," "summarize the top three concerns," and so on.
1. Sentiment Analysis
Available from the analysis tool on any open-text question. Gives you a bar chart showing the distribution of positive, neutral, and negative responses, plus a table showing the likelihood of each sentiment for individual responses. Responses in less common languages are automatically translated into English before sentiment analysis runs.
This is the fastest of the three - no setup, no prompts, just a chart. Use it as a first-pass pulse check, not as your final analysis: it tells you the tone of the room, not why people feel that way.
Read more: Analysis of open text questions
2. AI Tagging
AI Tagging turns large sets of open-text comments into a structured set of themes, without days of manual coding. Here's how it works:
- It reads all the comments once and picks out the 5-10 most common themes.
- It reads them again and tags each comment with the relevant themes.
Pasting everything into a general-purpose AI tool tends to cherry-pick and gives unreliable results. This structured process avoids that, and it's fully auditable: every tag, filter, and prompt is visible, so anyone can retrace your steps and land on the same result.
Read more: Tag open responses with AI
3. AI-powered summarization
Uses Claude to answer open-ended questions about your data: summarize responses, identify themes, or ask about a specific topic. Select the questions you want analyzed, click the AI button, enter a prompt (or use one of the defaults), and ask follow-up questions as needed. The tool only analyzes responses currently visible in the table, so refresh your selection if you change filters.
What actually gets sent to the LLM: The AI analysis tool in Maptionnaire works on a "What You See Is What You Get" (WYSIWYG) principle. It will only look at the data you have selected to be used in the analysis. Any maps, charts, or filters you currently have active in your analysis will be sent to the AI. This means the themes it highlights are always relevant to your specific focus.
This is the most flexible of the three - good for open-ended exploration or answering a question you didn't know to ask in advance. It's less structured than AI Tagging, so it's a better fit for a quick, specific question than for producing a citable set of themes for a formal report.
Here's how to do it: AI-powered analysis of responses
Dive Deeper: Use Filters to Compare Groups
You can easily check if opinions differ between groups (like age, neighborhood, or how long someone has lived there). Just use the filter tool to select responses from, say, "People who cycle to work." Then, let the AI summarize the open-ended comments only for that specific group. This gives you highly targeted and useful feedback.
Want to learn how to filter your data? Check out this guide: Filter charts
Use Spatial Filters to Focus on Interesting Trends on the Map
Notice that many respondents marked a specific location on the map, and want to see what that's all about? Use the Spatial Filter option to focus on that specific spot:
- Select all the questions in Charts that you'd like to use, e.g. demographic questions to see the profile of respondents who made these map comments, and any pop-up questions linked to the map question.
- Open Filters > Spatial Filters. Draw borders around the area that you want to take a look at. Hit "Apply".
- Now all the charts are filtered to show only responses given by the respondents who wrote about this location. You can see what their profile is (e.g. if they are small business owners or residents) and what they commented regarding this place.
-
Open the AI tool and ask it questions – e.g. ask it to summarize the open comments left related to the place.
About Claude and data
As of August 2026, Maptionnaire uses Claude Sonnet 4.6, running on Amazon Bedrock — the same AWS infrastructure that hosts the rest of Maptionnaire. The model does not retain your responses after processing them, and does not use them to train models. All AI processing of response data happens in the same AWS region as your data — EU data stays in the EU, US data in the US.