Ask HN: My attempt at empowering companies to make better product decisions
As a developer, I have built many features. I wouldn’t say all, but many of them were a waste of everyones time. These features were usually built because the product group told us to build them. I found product teams make these decisions based upon some customer interviews, email feedbacks, phone complains and mostly their “gut”. And its hurts the most when we would bust our asses building and releasing it, to only realize that our customers are not using it. Reasons vary from customers wanting X and we building Y to the user flow not being right. Its further disappointing to learn after a release that customers didn’t even care enough to click on the feature i built, forget about using it.
Prioritizing what to build, which features to double down on and which ones to ignore, what to prototype and what to scrap etc. are all important questions which the product team needs to decide. In their defense I feel like these are difficult questions. To me its just that I don't think they can afford to get this wrong. Building the wrong features, maintaining them and then eventually killing them is very expensive and just a moral killer.
The question in my mind is can product teams make better product decisions? Why do we have to ask developers to build things, release it and then learn whether customers really wanted it. To address this issue and empower product teams I have built www.featureKicker.com. It enables teams to deploy features before building them, collect feedback, analyze data and make better, more informed, data backed decisions. It enables a product manager to look a developer in his or her eye and say “I need you to build X and I need it done by Y. I have Z number of customers waiting for it. Here you can read what they have to say”.
I would love to hear from the community of what they think of this solution. I want to help companies make better product decisions and am looking at the community for guidance.
Sandeep
CTO
FeatureKicker.com I checked the website, looks promising so congratulations on that.
I do have a comment, your entire website is still linking to your Heroku subdomain. If I'm not mistaken, this pretty much feels like A/B testing with the implementation being a survey. So have you tested users' sentiments about trying to use a feature only to realize it isn't implemented and then being promoted to answer a few questions? I'd speak for myself, I'd rather not see a button to for example Authenticate with my Twitter account if it doesn't work, than to see one of which when I try to use it, I get asked a few questions. I'd feel that's a bummer. To that point, why would I use FeatureKicker instead of a service like Qualaroo, which works fairly well. Cofounder of FeatureKicker here... >> "I checked the website, looks promising so congratulations on that. I do have a comment, your entire website is still linking to your Heroku subdomain." I hear you. We were being as cheap as possible and didn't want to pay for SSL certificates, etc. while building early version of the product. But we're hearing this more and more, so perhaps it's time to move on :) >> "If I'm not mistaken, this pretty much feels like A/B testing with the implementation being a survey." It's interesting and refreshing to see someone else boil down our product. Thank you! So I agree that it "feels" like A/B testing because you're adding a new element (and potentially removing it based on our rules engine). But it's unlike A/B testing because our tool is not designed to split your traffic across page variations. Here's what I would add to your distillation: it's being able to ask the right user, the right question, at the right time. And this is where our product is like a "hyper targeted survey". >> "So have you tested users' sentiments about trying to use a feature only to realize it isn't implemented and then being promoted to answer a few questions? I'd speak for myself, I'd rather not see a button to for example Authenticate with my Twitter account if it doesn't work, than to see one of which when I try to use it, I get asked a few questions. I'd feel that's a bummer." Absolutely. This was our primary risk. After user testing, we're finding that this concern is more of a theoretical anxiety than actual feeling in practice. >> "To that point, why would I use FeatureKicker instead of a service like Qualaroo, which works fairly well." I think the tools and use cases are different. With Qualaroo, you get a pop-in question based on a timeout, and those questions are typically related to overall customer satisfaction or net promoter score. But we believe that using Qualaroo to ask a specific question about a specific feature on a page will not work as well as FeatureKicker. Why? Because the question may be irrelevant to whatever the user is doing at that time. This gets even trickier when you want to ask a question about an unbuilt feature. Then you have to worry about showing the experimental feature only sometimes and coordinating your Qualaroo question to pop-in when you're selectively displaying the experimental feature. In contrast, FeatureKicker allows you to ask the right user, the right question, at the right time. Let me unpack that. It's the right user because it's the person using a particular (built or unbuilt) feature. It's the right question because you're going to ask something relevant, specific about that feature. And it's the right time, because you're capturing the user at the point of interaction, which is the peak of their curiosity. We believe this explains why we're seeing up to 64% response rates to our clients' questions. I think this is amazing.. a huge problem all around and a very clever solution. Surveys are broken, putting the feature on site and testing who clicks on it-- so good!! Engineers all over the world thank you!! Any thoughts on how to do this for "non-button" content features? Like say, do our users want to see product images in the search result page? GovindKabra: Thanks for your feedback. Recently my co-founder asked me to build a google map for a feature showing all thumbs up it received displayed by location. Rather than building this feature, I kicked it on our page. We also have javaScript API's which can be used to show an overlay.
Here is an example of it being used after a customer does a search: "It asks customers, do you like the search results? and can we improve our search feature?" I have one beta customer showing overlay after 90 seconds have passed on the "buy-now" page and another beta customer showing overlay when anything is clicked besides the "buy" button. I hope these examples show how beta customers are using FeatureKicker to solicit feedback. Amazing, amazing, wow this is so good. Lots of astroturfing here even from very new accounts (not yours), why so many exclamation marks with praise when the real value of HN is in getting you actionable feedback? Jacquesm: Did you have any feedback on featurekicker or in general about making better product decisions? Would luv to get your insight. Update video to show g+ example and results no more no less about 20 secs tops - the video on main site takes to long to show the meat of what you do... bar that, not a bad product we just built something not to dissimilar in house for our needs, this would have saved a ton of time... Good luck with the product. -Scott Awesome and timely feedback. Thanks, Scott. I signed up for your beta. Definitely seems like a wonderful idea. Looking forward to trying it!