I built the feedback tool I wanted as a user, and needed as a founder
I believe collecting user feedback is one of the most important things you can do while building a product. I used to run a B2C app with 100k monthly users a couple years ago, and got so frustrated with existing solutions which were either too complicated or too expensive, that I ended up using a single Google Form. I don't need to tell you how tedious the process of going through all the answers was.
Fast forward to today, I sold that app, and for the last 5 months I've been working on Modu.io , a feedback collection tool that allows businesses and communities to create multiple kinds of feedback modules (suggestions with voting, roadmaps, changelogs, polls, ratings, open questions) and either organize them in a public board, link to them directly, or use them as in-app embeds/popups.
Other than stressing a lot about how the modules look, I've been also working on the behind the scenes to make it easy to analyze the collected feedback. Other than integrating with all major tools (jira, clickup, slack, trello, google sheets, linear), Modu also automatically clusters text feedback, grouping all similar answers to a form, detects duplicates on public suggestions boards, and notifies you when important targets are met (e.g a suggestions reaches 10 upvotes, a rating poll average score changes, etc.).
The tool is highly customizable, both in looks (colors, logo, favicon, style) and in how you organize your boards, so I'm really excited to see how people might use it :)
I really like how Modu focuses not only on collecting feedback, but also on making it actionable through automatic clustering, duplicate detection, and integrations with tools like Jira, Linear, and Slack. That shifts the value from “yet another form” to an actual decision-support layer for product teams.
Curious about the duplicate detection on public boards - does it work purely on text similarity or does it factor in context? I've seen tools flag "this doesn't work" and "X feature is broken" as different issues when they're clearly the same problem.
The credibility of actually managing 100k users with Google Forms before building something is huge. Most founders build the tool first and hope people need it. You lived the pain. Curious though, what's the biggest friction point when getting teams to switch from their existing feedback setup? That adoption hurdle from 'we already have a process' to 'this is genuinely better' seems like the real challenge, especially when the current workflow technically works even if it's painful.
“built the tool I wanted as a user” is exactly the kind of origin story that usually leads to sharp product decisions. The pain of scattered feedback across tools and docs is so real. I’m building Zyvia (task management for indie founders) for a similar reason: Jira and Linear feel broken for small teams that just want to move fast without wrestling a heavyweight system. Really curious to see how Modu evolves once more teams plug it into their real workflows.
The lofi.co origin story adds so much credibility here. You've actually lived the pain of 100k users and a Google Form — that context makes Modu feel like it was built out of necessity, not just opportunity. The auto-clustering is the real differentiator. Most feedback tools dump raw data on you and expect you to find patterns. Sounds like you're solving the 'drowning in feedback' problem, not just the 'collecting feedback' one.
The lofi.co background gives this a lot of credibility. You've actually dealt with 100k+ users worth of feedback — that's not something most feedback tool founders can claim. The auto-clustering is what would sell me honestly. I've tried Canny before and the biggest problem wasn't collecting feedback, it was drowning in it. You get 50 variations of "make it faster" and 30 ways people describe the same bug, and you end up spending more time categorizing than actually building fixes. If the clustering actually works well with semantic similarity (not just keyword matching), that alone is worth paying for. One suggestion: the notification triggers you mentioned (suggestion hits 10 upvotes, rating average changes) are cool but I'd want Slack alerts tied to sentiment shifts, not just thresholds. Like, "hey, 5 people mentioned checkout issues in the last 24 hours" is way more actionable than "suggestion X hit N votes." That's the kind of signal that helps you catch fires early before they show up in churn numbers.
Love the story — building a tool to solve your own pain is usually how the best products start.
The auto-clustering and integration approach sounds super smart — that “turning raw feedback into actionable insights” step is where most tools fail.
When I ran my own projects, the biggest game-changer was moving from manually reading responses to having something that automatically grouped similar issues — it saved hours every week.
For someone starting out: focus on the feedback loop, not just collection. Ship something small, learn from real user patterns, and iterate quickly. The speed of learning beats perfection any day.
Curious — are you seeing more insights come from power users or casual ones? In my experience, the casual users are the silent goldmine.
Love that you built this from actual pain — the Google Form → manual analysis workflow is so tedious. The auto-clustering feature is the real differentiator here. Most tools dump data on you, but making sense of it is where the value is. Smart integrations too. Congrats on shipping!
I like the origin story here, solving your own pain point often leads to the most practical tools. I recently adapted a traditional blessing for a family birthday and it felt sincere, much like the warmth in happy birthday wishes for friend in marathi from birthdaywishesmarathi .net. A helpful next step could be sharing a brief case example of how reduced feedback processing time, since concrete outcomes make the value immediately clear.
Great idea! How is it now? Are things improving and what are some advice for someone like me who wants to get started as well?
Cool concept! The gap between what users say and what they actually do is real. Especially with your first 100 paid users. One thing we noticed: in-app feedback catches the users who bother to write. But the ones who churn silently? Those are the goldmine. For that we use Saasfeedback.ai. They actually call users after churn/trial drops and give you the real reasons + feature recs. Different approach (warm calling vs widget) but complementary. Curious: are you seeing more feedback from power users or casual ones? In my experience, the casual users are harder to reach but have the most valuable insights. I'm really trying to focus on the "silent majority"
I am in the same boat that you are in. I use google form - actually an airtable link and collect feedback. And I cant for the life of me spot patterns, so i download the file, chuck it into claude and ask it to give me analysis of nice things people are saying / bad things that i should look at and give me a list that i need to upload into the task list to fix / rethink.
The UI for your product is beautiful. I have to think about how it would fit into my site (you mentioned it allows for it), but will definitely sign up for this and play with it. Thank you for sharing.
Thank you! If I recall correctly, at some point I tried airtable too but didn't really like the way it showed answers. Not that Google Form is that much better, but I preferred it. Feel free to shoot me a message if you need any help with the tool!
I think Airtable is superpowerful. Their UI masks the fact that it can be quite powerful. Excel is ugly and smart. Airtable, because it is pretty, people expect it to be behave like canva, but it has photoshop level capabilities. Most folks just end up using it like a form. How many tools can one person master anyways :D
The "Google Form + manually reading 1000 responses" pain is so real. I've been there.
What stands out here is that you're solving the entire feedback loop - not just collection, but analysis and action (via integrations). Most tools stop at "here's a form, good luck."
The auto-clustering feature is interesting.
How does it handle edge cases? For example:
- "Your app is too slow" vs "Loading takes forever" = same issue, different words
- "Add dark mode" vs "My eyes hurt at night" = related but not identical
Does it use embeddings/semantic similarity, or keyword matching?
Also curious about the business model:
You mentioned existing tools were "too expensive." What's your pricing strategy? Are you going after the same market (enterprise) with better pricing, or targeting smaller teams/indie founders who got priced out by tools like Canny/UserVoice?
The integrations list (Jira, ClickUp, Linear, etc.) suggests you're going after product teams specifically. Smart move - they have budget and feel this pain daily.
Congrats on shipping! The "built it because I needed it" products tend to have the clearest vision.
Really like the black and green design of the website. Clean, focused, and product-driven. And building your own feedback system after running a 100k MAU app makes a lot of sense.
Quick question from a trust and integrity perspective. How are you verifying that votes, ratings, and suggestions are coming from real users and not bots or scripted abuse? With public boards and in-app embeds, feedback manipulation can quietly distort product decisions.
Are you using identity validation, rate limiting, or abuse detection behind the scenes to protect signal quality?
Curious how you are thinking about protecting the feedback layer as adoption grows.
Thanks!
Other than built-in rate limiting and spam prevention, tenants can lock all modules to verified only users (normal signup, google oauth or SSO), or even password-protected specific boards.
Love that approach 👍. Giving tenants control over locking boards to verified users and SSO is a strong foundation.
One thing we often see as products grow is that manipulation doesn’t always look like obvious spam. It can be coordinated low volume voting, scripted “real looking” accounts, or internal abuse from shared environments.
At Nautillo Pro we focus a lot on simulating how seemingly protected features behave under pressure, especially where user input influences business decisions.
Out of curiosity, are you logging anomaly patterns across tenants, like unusual voting velocity or repeated behavior from similar device fingerprints, or is protection mostly per board level right now?
Congrats on the launch, looks solid. How are you currently thinking about acquiring early users and gathering feedback?
Thanks! Currently posting on reddit and on the usual suspects, including the main launchpads (except ProductHunt, for which I'll wait for traction to start coming in). Started SEO as well, but that will most likely take a while to kick in
That makes sense, Reddit can be incredibly powerful in the early stages, especially for gathering feedback if used effectively.
The challenging part is that many founders tend to post in the more obvious subreddits and overlook the smaller, high-intent threads where people are actively discussing specific pain points. This is usually where real traction can be gained.
Are you currently focusing on broad visibility, or are you trying to spark conversations within specific niche communities?
This comment was deleted 6 months ago