WhatsApp Wrapped tools are fun. You upload your chat export and get colorful stats: message counts, busiest hours, most-used emojis, longest streak.
ThreadRecap does something different. It reads the conversation and tells you what was actually said.
Typical Wrapped-style outputs include:
See what ThreadRecap finds in your own chat.
Analyze your chatThese are stats — similar to what a WhatsApp Wrapped tool gives you. They tell you how much you chatted, not what you chatted about.
Statistics describe behaviour around a conversation, not the conversation itself. Knowing that your group chat peaked at 9 PM on Thursdays tells you nothing about the project decision that was buried in that flurry of messages. Knowing that one participant sent 40% of all messages tells you nothing about whether their messages contained commitments, questions, or complaints. Wrapped-style tools are built on metadata — timestamps, participant identifiers, character counts — which means they can only ever surface patterns in how people communicate, not substance of what was communicated. That is not a flaw in those tools; it is simply the scope they are designed for.
When you export a WhatsApp chat, the resulting ZIP contains a plain-text `.txt` file and, optionally, attached media. The `.txt` file holds every message in a structured format: timestamp, sender name, and message body. Voice messages are stored as `.opus` or `.m4a` audio files and are referenced in the text file by filename. Wrapped tools parse the text file to count and categorise messages. They note when a voice message filename appears but treat it as an opaque attachment — the audio content is never processed. This is why voice message counts appear in Wrapped outputs while the spoken content remains invisible.
ThreadRecap outputs include:
These are insights. They tell you what happened in the conversation.
The structured summary ThreadRecap produces is not a simple concatenation of messages. It groups related exchanges into topics, preserves chronological context for each decision, and surfaces action items with the name of the person who agreed to do something and, where mentioned, the deadline they committed to. Open questions — things that were raised but never resolved in the thread — are listed separately, which is useful for follow-up. Notable quotes are pulled verbatim from the conversation, so the output stays grounded in what was actually said rather than an interpretation of it.
ThreadRecap supports exports containing 75,000 or more messages and ZIP files up to 2 GB. In practice this covers multi-year group chats and high-volume work channels that would be impractical to read manually. Wrapped tools also handle large exports, but the processing demand is lower because they are counting rather than comprehending. For ThreadRecap, the ability to process that volume means you can run it on an entire year of a busy project channel and get a coherent summary, not a timeout or a partial result.
Use a Wrapped tool when you want:
Use ThreadRecap when you want:
Personal chats and group social threads are exactly where Wrapped tools shine. The stats are relatable, the graphics are shareable, and the emotional value comes from the numbers themselves — "we sent 12,000 messages this year" is a conversation starter. Work chats, project channels, client threads, and family planning groups are where ThreadRecap earns its place. These conversations contain commitments, plans, disagreements, and decisions that have consequences. Counting messages in those threads is less valuable than extracting what was agreed and who agreed to it.
Yes. They solve different problems.
Run a Wrapped tool for the fun stats. Run ThreadRecap for the actual content analysis. There is no overlap because they extract completely different things from the same data.
The fact that both tools take the same WhatsApp export as input does not mean they compete. A Wrapped tool will tell you that your friend sent 3,200 messages last year. ThreadRecap will tell you what those messages were about, which decisions came out of them, and what questions were never answered. Running both on the same export gives you the full picture: the behavioural pattern and the substantive content.
Most Wrapped tools ignore voice messages entirely. They count them ("47 voice messages sent") but do not listen to them.
ThreadRecap transcribes every voice message and includes the content in the analysis. If someone made a commitment in a voice message, it shows up in the action items.
ThreadRecap uses OpenAI's transcription models for voice messages, processing `.opus` and `.m4a` files — the two formats WhatsApp uses depending on device and platform. Clear audio transcribes best; noise, crosstalk and strong accents degrade the result. The transcribed text is inserted into the conversation timeline at the correct timestamp, so the analysis treats voice messages and text messages as a single coherent thread rather than separating them. This matters because in many active group chats, critical information is conveyed via voice message precisely because it is faster to speak than to type. Leaving those messages unread means any summary is incomplete.
Consider a scenario where a project decision is discussed over text messages, then one participant sends a 90-second voice message summarising the agreed next steps. A Wrapped tool records that one voice message was sent. ThreadRecap transcribes it, identifies the next steps as action items, assigns them to the speaker, and includes them in the output. The difference in usefulness is significant — one output tells you a voice message existed, the other tells you what it said and what it means for your project.
Wrapped tools show who sent the most messages. ThreadRecap shows who made the most decisions, who has the most open action items, and what topics each participant focused on.
For work groups, the chat analyzer output is actionable. The Wrapped output is interesting but not useful.
In a group chat with five or more participants, people tend to specialise. One person raises blockers, another proposes solutions, another asks clarifying questions. ThreadRecap's per-participant topic analysis surfaces these patterns explicitly. You can see that one team member consistently drove scheduling decisions while another owned technical discussions. This kind of breakdown is not available from Wrapped tools because it requires understanding message content, not just counting messages. For managers reviewing a project channel, or clients trying to understand team dynamics from a shared thread, this is information that message counts cannot provide.
For group chats specifically, ThreadRecap identifies who proposed decisions, who confirmed them, and who was assigned follow-up tasks. This is directly useful for accountability. If a deadline was agreed in a chat and later disputed, the ThreadRecap output shows the exact message, the participant who agreed, and the timestamp. Wrapped tools have no mechanism for this because they do not parse the semantic content of messages.
Both types of tools require your WhatsApp export. The difference is what they do with it:
When ThreadRecap unzips your export locally in the browser, the ZIP file is decompressed using JavaScript running on your device. The photo, video, sticker, and document files in the export are never transmitted to any server. What is sent for AI processing is the text content of the chat, the voice messages you choose to include, and, if you include videos, the audio track your browser extracts from them. This architecture means that for a typical personal or family chat, the vast majority of the export — which is usually dominated by photos and videos by file size — stays entirely on your device. For sensitive work conversations, this distinction matters: confidential documents shared in the chat are not part of what leaves the browser.
Wrapped tools answer "how much did we chat?"
ThreadRecap answers "what did we talk about and what do we need to do?"
If you need the second answer, try ThreadRecap. Upload your chat now.
Upload an export and judge the results on your real conversation, with free credits to start.
WhatsApp AI summaries catch you up on unread messages, while ThreadRecap exports create detailed recaps with full context and customization.
Jan 31, 20266 min read
ThreadRecap turns your WhatsApp export into real insights, not just stats. See how it compares to Wrapped tools and find out which one you actually need.