Launch HN: 1Flow (YC W22) – In-product user feedback for web/mobile apps (1flow.app)
It’s really hard to get reliable feedback from users. We solve that problem. Our surveys have a 50-70% response rate on average, compared to 5-10% for traditional surveys.
I studied product design at college and then joined a mobile gaming company. It was painful to see how far removed product teams are from their users in the real world. We didn’t talk to users directly, as my design program had taught me. Instead, we relied on monthly reports from customer support and made top down decisions on what to build.
Then I started a mobile app company of my own and also experienced this problem—it is actually very hard to reach users and hear their thoughts. We didn’t have a good way of contacting them, or often even know who was downloading our app. And when we did manage to contact them, they wouldn’t respond. I tried showing a pop-up in my app that led users to a Google form. I was surprised / frustrated when after 2 weeks of waiting, I received zero responses.
But then when I was doing ASO (aka SEO for the App Store), I got a lot of users to write me good reviews after I had implemented a bottom-drawer UI that asked them to please leave a review if they liked us. This UI was only a half-page in size, showing from the bottom edge of the screen (close to the thumb) and the rest of the screen was only darkened, not fully blocked. This worked like a charm! I was able to get consistent reviews from my users on a daily basis (when before I was getting 1-2 reviews per month).
I connected the dots and realized, what if I could do user surveys effectively by giving people a user experience that they actually like? And went to work.
1Flow embeds a JS code snippet in websites / web apps, and a native mobile library in mobile apps. After this initial 5min setup, anyone—technical or not—can use our cloud dashboard to create and launch new in-app surveys that ask users about feedback, satisfaction, and opinions. No app updates or code changes are required beyond the initial installation.
Our in-app surveys can easily be designed to match the branding of the client application, so app makers can engage users for feedback in key moments of the user flow without compromising their user experience. Results from the surveys are available on our dashboard in real-time. We also have filters that allow app makers to view results by cohorts, location, and answers to a particular question.
We provide webhooks and csv download for teams to easily feed data into other tools they are using (Amplitude, Slack, GA, Intercom, Hubspot) - so they have a single source of truth for qualitative and quantitative data.
I've met developers who desperately needed a solution like this, went into code and hacked together a question flow inside of their apps. But then they are stuck with this hardcoded questionnaire in their code base. Every time they need to update it they need to spend engineering hours to go inside the code and update it, then for mobile apps they'll have to submit the new version to the App Store / Play Store and wait until the updates are approved. Also, it gets complicated to target certain users only (e.g. users in France who completed 3 transactions recently) and automatically stop collecting responses, e.g. after 500 people have given their answers. Visualizing the data and filtering/slicing it is also really hard if you don't have a dedicated tool.
There are other embeddable survey tools, but we’re different in three ways. First, most of these tools have poor UI/UX, so much that it's painful for developers to put them inside their beautifully crafted apps. Second, they tend to offer static forms that can't be used to target certain users based on their app behaviors, and can't be used to enrich your CRM / analytics easily. They are also a pain to manage....
65 comments
[ 4.8 ms ] story [ 150 ms ] threadThis is what I currently have for https://nocommandline.com (but it’s positioned at bottom right of the page) and just like your experience, it doesn’t seem to have worked.
Will give your App a spin.
Actually the bottom-drawer UI is now the default for Android in-app store review API:
https://developer.android.com/guide/playcore/in-app-review
Not to be nitpicking, but it’s not (only) the cookie that requires consent, but the storage and the use of the data itself. So if your customers do not have a cookie consent yet, they might even have _more_ work to do to implement your solution because they still have to ask for permission and store the answer.
But this takes it to the next level.
<3 Feedback like this make all the effort of doing talks worth it!
> If you or someone else you know would benefit from using our solution, we'd love to chat!
I'm messaging you right now!
Note that higher response rates don't always translate to better data. As you make the survey experience more intrusive you incentivize certain users to provide junk answers simply to remove the barrier. Just something to consider... especially if you decide to scale to enterprise.
Congrats on the launch! It’s great to see other companies emerge in this space. Product managers, designers, researchers too often rely on panels with people placed in hypothetical situations. Research is most valuable when conducted with actual customers as they are experiencing your product.
We haven’t seen the problem as you’re describing for smaller startups. It’s actually better to speak with customers directly 1:1 until your product starts to achieve scale.
One suggestion for you is to add video questions. This is a great way for early-stage startups to see and hear from users directly. It’s been so well received by startup founders that video questions are included in Sprig’s Free plan.
Also, the 50-70% response rate is suspect. Sprig has surpassed 2 Billion unique users tracked and millions of survey responses for customers including Dropbox, Loom, and Square. We’ve seen response rates as high as 90%, but on average are seeing a 30% response rate. 1Flow’s survey design is an exact clone of Sprig (see comparison: https://www.loom.com/i/356c650a70b94fffa9a85da83b546595) so differences in design won’t be a factor. Even a 30% response rate is significantly higher than an email survey though which is actually around 2-5%.
I've actually talked to a lot of startup founders, product leaders, and even current customers of Sprig, and learned that most didn't want to put video chats inside of their app because of how disruptive it is to the user experience. Zoom, UserTesting.com, etc. have a lot better ways of doing this and they've been doing it for years successfully. We think you're really serving big brands user research teams well because they really need video customer chats and because of our different approach to who we serve and our design philosophy, we don't yet see it a priority to add video.
We did months of customer research before building 1Flow - if your users are truly happy, we wouldn't exist.
With regards to the UI "clone" issue, I couldn't agree. There are already many tools such as Pendo, Appcues, Survicate, etc. that are using this approach, but as I had explained in my post, it is really about providing an experience both software makers and their users will love - at least that is the goal of 1Flow. Thank you for brining this to our attention, with regard to UI, I think we definitely can do a better job! :)
Our response rates are based on true data we see. We are a smaller startup trying to serve other startups of the world, and we are not serving enterprise customers at our stage. So I couldn't join you in making this a number competition, and also not interested in. All I can say is your 90% seems one-off, but I understand how things work and wouldn't want to take you up on this.
Finally I want to say that we are both trying to innovate in a space traditionally dominated by players like Qualtrics, SurveyMonkey, Medallia, InMoment, and 999 other survey tools. So I'd LOVE to stay connected with you and support each other however we can.
Kai
P.S. As founders we are all bit scared about competition, I understand. At 1Flow, we've tried our best to focus on actually delivering value to our users.
- AirBnB wasn't the first home sharing site
- Stripe wasn't the first payment platform
- Facebook wasn't the first social network
What really matters at the end of the day is finding product market fit and execute well.. This is just my 2 cents.
Disclaimer: I'm the Founder/CEO. Just send me a note (see my profile) and I'm happy to get you setup or you can create a free account on our website.
Because this is a YC startup's launch thread, I would normally hesitate to post like this (we moderate less when a YC startup is involved: https://hn.algolia.com/?dateRange=all&page=0&prefix=false&qu...), but I would say (and have said) the same thing in non-YC launch threads, and something about this case feels worse than usual.
Launch threads are a bit different from regular threads in this respect. It's of course fine for users to sincerely ask how the launching thing is different from existing things; it's borderline for a competitor to post a link to their thing, depending on how they do it; but to try to divert discussion to one's own thing is just bad manners.
I've detached this subthread from https://news.ycombinator.com/item?id=30375868 and marked it off topic.
For me, I work on personal side projects for weeks if not months and need to gather feedback from my loyal fans (family) - I would love to see a free option for 100 or less users - forever.
1. What comes with the free tier? Says I can get started for free but not sure if it is a trial or dev account etc.
2. I would love a way for the widget to close if the user clicked outside of the popup instead of having to click the X. It would be nice to have this as a setting to be less intrusive on less important surveys.
Actually just thought of a third.
The steps seem to load rather slow (well, not instant) with a fade effect. Totally down with a quick fade effect but would like it to be a bit snappier, not sure if you have to query for the next step but if you are, and thats whats causing the slowdown, maybe pre-fetch the next step (could be on hover of an answer if there are a lot of branches), or first step before showing the widget, so it can be a bit faster.
1. Free tier allows you to test for several days without limits and no need to put down CC / payment.
2. Widget to lose when clicked outside: I like this, have put this down on our list. This is something that I'd personally want to see using 1Flow as well!
3. Yeah I agree. Animation should be faster, especially because we load everything instantly so it's just the frontend animation taking time.
Thanks for your feedback! Cool ideas and very practical improvements we could make.
Would love to see a developer tier. We aren't launched yet so it's always hard to integrate something into the product when you don't exactly know when that will be and have to pay in the meantime. Even if it's EXTREMELY limited (10 tracked users or something) as long as I don't get kicked out I would be happy.
I saw that "...works 100% identical in production and development mode, except that we don't backup the data from development apps nearly as often" - so does this mean you don't really prevent users from using it in prod if they want to, but just warm them that this may result in loss of data and therefore is not recommended?
Only supporting snippets signals to me that a company might be behind the times and could be dealing with globals in a way that would cause problems in a complex web app.
Having an npm module as an integration option is a major selling point to me, and I think it might benefit you guys to call this out more clearly.
Regardless, congrats on the launch.
Feedback like this is what makes HN great. Thank you!
1) - Slow fade in is annoying and is just a lag on inputting the feedback.
2) - From what I can tell it doesn't move to the next 'slide' until it receives a response from the server which again feels really slow, if possible move it to the next slide immediately whilst it's sending messages to / from the server in the background.
Great looking product though!
* Also have a dedicated email input field if capturing emails is a value add, having it in a text box that has a 280 character max looks strange