Opportunities › Text classification for qualitative research › Proposed app
Themewise
Sort thousands of survey answers into your own themes in seconds, no training needed
- Score
- 34
- confidence 46
- Evidence
- 5
- items on the page
- Rivals
- 1
- named in sources
- First version
- 25 d
- estimate
- Type
- Desktop app
- PWA: runs in the browser, installs on Mac and Windows
- Market since
- 2023
- Emerging market · AI estimate
ScreensConcept
Desktop app: Researchers work with large datasets, define themes, and review classifications in long work sessions that require detailed table views and keyboard input. Sample screens on a computer, in the proposed brand. Their numbers are sample data.
New project
Ready to run4,812
About this appAI proposal
Themewise lets qualitative researchers sort large sets of open-text responses into themes they define in plain words. Paste or upload a column of answers, name your themes, and each entry is tagged. No training data, no code, no model setup.
It matches meaning, not keywords. If one answer says pasta and another says penne, both land under Italian food. Similar themes are kept apart, and every tag shows a confidence level and the words that led to it, so you can check the work.
Unsure tags go to a review queue instead of being buried in the totals. You fix a few, refine the theme wording, and re-run the whole set. Export tagged sheets for your analysis tools or your report.
- Tagline
- Your themes. Thousands of answers. Sorted in seconds.
- Voice
- ClearCalmPractical
- Why this name
- Theme plus wise: a tool that understands what responses mean. Two plain words joined, easy to say and spell.
In the first version
Members say which features they would use. It tells PWAS what people want; it changes no score.
Themes in plain words
Describe each theme in a sentence and add optional examples. No labelled data or training step is needed.
No member yet
Meaning-based matching
Recognises related ideas, such as pasta and penne under Italian food, rather than relying on exact keywords.
No member yet
Close themes stay apart
Add a contrast note so near neighbours, like cost worries and price praise, are not mixed together.
No member yet
Confidence and evidence
Each tag shows a confidence level and the phrase behind it, so you can judge it quickly.
No member yet
Brand identityAI proposal
Check the name against trademark registers before using it.
Palette
#12525ADeep Teal · primary#FAF6EEPaper · surface#1D2527Ink · text#E8A33DSaffron · accent#C4573AClay · data#8DB19ASage · data
Typography
Aa
Themewise
ABCDEFGHIJKLMNOPQRSTUVWXYZ 0123456789
PricingAI proposal
The material gives no competitor prices, only a stated wish for cost under 0.1 cents per classification. These tiers are a starting point to test against that target.
Free
Free
- 500 entries per month
- Up to 5 themes
- CSV import and export
Researcher
$19 a month
- 25,000 entries per month
- Unlimited themes
- Review queue
- Shareable summary pages
Team
$59 a month
- 150,000 entries per month
- Shared projects
- Theme libraries
- Priority support
Would you pay Researcher · $19 a month?
Your answer tells PWAS which prices people accept. It changes no score.
No answers yet.
What it could be worthEstimate
What a software business making this much sells for today, from real deals. Pick a monthly revenue.
If it makes
Worth about
$194,000
Sale range $132,000 to $263,000
$120,000 a year in revenue
From 87 software businesses listed and sold on Empire Flippers in the last year, revenue vetted by the broker: they sell for about 19× their monthly revenue, most between 13× and 26×.
Measured on profit, the middle deal is 32× monthly net profit: about 2.7 years of profit.
An estimate, not an offer or financial advice. A real price depends on profit, growth, churn, how much of the work is the owner's, and the buyer.
How it would growAI proposal
How the product would reach new customers through its own use, and how the first ones would be found. A plan to test, not a result.
- Built-in loop
- Shared outputs. Researchers publish read-only theme summary pages and attach exports to reports, papers or team updates. Each page carries a small made-with note linking to the tool.
- Do they become customers?
- Readers are often fellow researchers who could try the free tier. Test whether summary page viewers click through and run their own project, not just view it.
- First customers
- Solo UX and market researchers at small teams who run open-ended surveys of 1,000 to 20,000 responses and have no data scientist.
- Where to reach them
- Hacker News and Reddit: Share a worked example sorting a public survey dataset, inviting researchers to try their own.
- UX research communities: Offer free runs in research Slack groups and forums in return for feedback on accuracy.
- Short how-to content: Publish guides on coding open-text survey answers faster, linking to the free tier.
- First milestone
- Five paying researchers in one vertical, each running at least two projects of over 1,000 entries.
Why people already pay for this
- VerifiedHacker News post received 13 points and 20 comments
- Public claimJev allowed non-technical researchers to classify arbitrary participant entries without prior training at near instant speed
- Public claimThe tool could classify thousands of participant entries as mentioning a specific topic with high confidence and differentiate inversely
- Public claimNon-technical researchers find this type of tool helpful in qualitative research
- Public claim
Tech stack
Every proposal builds on the same PWA base, installable from the browser on any phone or computer. Dashed boxes are what this app adds.
- AppNext.js and TypeScript
- PWAManifest and service worker
- UITailwind CSS
- Data and authSupabase Postgres
- PaymentsStripe
- EmailResend
- HostingVercel
- This app · Language model accessHosted LLM API with batch supportZero-shot, meaning-based classification of entries without training data.
- This app · Background jobsQueue worker for batch runsLarge batches need queued processing with progress and retries.
RisksAI inference
- AIClassification quality was disputed by commenters on the source post. Accuracy on subtle themes must be measured against researcher-coded samples before claims are made.
- CommercialA sub 0.1 cent cost per entry leaves thin margins and invites free alternatives, including researchers scripting a model themselves.
- DataParticipant responses may be sensitive or covered by ethics approvals; researchers need clear data handling and deletion terms.
- TechnicalKeeping cost low at thousands of entries needs batching and pre-filtering, which may reduce accuracy on close themes.
- CommercialAcademic buyers often have small budgets and slow purchasing, which may delay paid conversion.
Everything marked AI proposal or AI inference was written by a model from the evidence on the opportunity page. The proposed name must be checked against trademark registers before it is used.