New product features nearly always fail for the same reason; they're not popular enough. At a larger company, this may just hinder the growth of the company, but at a small or new company, this type of failure can actually sink the company.
In my experience (which is mostly startups) there's nearly always a similar issue at the heart of these failures, which is that people didn't plan on how to get new users, and they didn't have a plan to transition initial users into regular users of their products. Many companies have a very If You Build it They Will Come attitude, which might have worked in Wayne's World 2, but doesn't usually work for products. In fact, I've only heard of it working once, which was at Funny or Die (I didn't work there, but I knew some people who did.) Thing is, the reason Funny or Die was able to get so many users off the bat, is they had a celebrity endorsement when they launched which drove crazy traffic to their site.
Most startups don't have celebrity endorsement. So, they need a plan.
One of the most irritating things about companies that don't have a plan, is that people who should be out getting users will often start micromanaging the developers. At this point, I think I have burned months of my life arguing with non-developers about SQL databases, because non-devs often think that making sure everything technical is "perfect" is the highest priority. They assume that the best way to do that is to stay on top of the devs, and be involved in all the decision making. However, not only does this tend to slow the developers down, it often leads to worse technical decisions as the devs have to negotiate technical choices with people who don't understand them.
Even worse, however, is that they're not just wasting dev time, they're not doing important work that needs to get done! I hate to admit this as a dev, but honestly, you can usually find an adequate dev to get your job done. What is much harder to find, and what usually determines the success of a product, is the people who understand the flow between marketing and user acquisition.
Often companies will not hire sales or marketing people (or, "user development people" to use Steve Blank's term) until after there is a MVP of the product. The problem is, you don't really know what type of product to build unless you have sales and marketing people. A PM can't stand in for for sales and marketing (unless they're a hybrid PM who also does sales/marketing) because how you source your users will change the structure of your application. And, only the people on the ground actually sourcing users will deeply understand this.
User conversion from a consumer app that is planning on funneling users in from Facebook ads will function differently to an enterprise app where most of the users have established relationships with one of the sales representatives in house. Not only that, but it's not always apparent from the get-go which sales/marketing approach is going to work from the beginning, so you may need to experiment with that until you get it right. If you wait until after you've shipped a product to start your sales and marketing, you may find that you've wasted time building an application that isn't well structured for your most effective marketing plan.
Now, if you have enough time and money, this is fixable – but, time and money will end up being the ultimate failure points for any project; you always want to be conserving both these resources as much as you can.
One of the more successful strategies I've seen in B2B (as luck would have it, at my ex boyfriend's company) is to actually start selling features before you've built them. Of course, you need your sales team to have a pretty solid sense of what's possible dev wise so they don't promise anything insane, but if you do this, you know that what you build will sell, especially if you've already collected some money for it.
Often, the success of B2B really hinges on the kind of relationship you have with your customers, and how much power within the organization your contact has. Although I wish it weren't the case, it is often more effective sales-wise to have someone's boss in the client company champion your product than it is to have a well designed product. Have you ever noticed that much enterprise software tends to be less usable than consumer software? That's why. Companies are not as heavily penalized for low usability in the enterprise space if their sales teams develop a strong enough relationships with the right customer representatives.
The consumer space, on the other hand, is much more judgemental. Often, consumers are looking at what you've actually made, and there is usually no way around the fact that you have to build features that may not thrive when released to the general public. The best way around this is to build something that you are planning on changing, and prepare your team for the unsavory possibility that they may ultimately need to rip out some of their beloved features down the line.
One of the ways I tend to see consumer startups go south, is that as they're iterating to find market fit, the complexity of the application becomes unwieldily. People become attached to certain features that fail to perform, are unwilling to cut them. So each new feature becomes harder and harder to build as the codebase must be modified to accommodate old experiments that didn't pan out. Iteration will begin to slow down, so that if you haven't found market fit early, it becomes increasingly more difficult as time goes on because your iteration cycle is slower.
(If you do find market fit early on, a slowing iteration cycle is not a problem because you're growing your customer base and getting money; before you have customers, though, you need to be iterating as fast as possible.)
Another common mistake that consumer startups tend to make early on, is they don't recognize signs of success because they are often less spectacular than what they want them to be. For instance, at a women's fashion startup I worked at, of the original 300 people who were invited to use the application, only 1 or 2 of them made purchases. While it seemed disappointing at the time, this company actually went on to get tens of thousands of customers (it ultimately went out of business a few years after I left – why it went under is an interesting topic for another time, but from what I was able to gather, it had a consumer growth strategy that worked at an early stage, but wasn't enough to scale it to millions of users.)
On the more "opportunity missed" side, was when I was working at a video streaming company, we were having real trouble getting our users to watch our videos. However, we hired a bunch of people through Mechanical Turk to double check the trailers for each of the videos, and some of the trailer checkers ended up watching the movies on the site. The PM at the time made an offhand remark about it to me, and about how it was kind of funny.
And, I missed the boat on this, but in retrospect I realized we should have capitalized on those users. Even if we ended up getting our initial customers through Mechanical Turk, would it have mattered? The problem was, we didn't think of Mechanical Turk as a viable user acquisition strategy. We thought of it as a strategy for proofing our content; the fact that it was actually more successful for user acquisition than our fb/google ads was something we simply ignored.
Now, I'm not saying we should have stayed with Mechanical Turk forever, but we could have used initial users from that service to design our system better. We had learned at least one thing already; getting people to watch the trailers could lead to sales. Maybe if we kept going with the Mechanical Turk users, we could have learned other things that would have helped us design better user acquisition strategies on other platforms.
Thing is, when you're at a startup, you can't be snobby if your users come from the wrong place, or if they behave in the wrong way. You have to take whoever you can get, and learn as much as them as possible. And, it is this particular thing that people tend to have trouble with; they either ignore the users they are able to get in hopes of getting ones that looked more like their "ideal" users, or they don't know how to pivot to accommodate the fact that people aren't using their application the way they initially intended.
At the base of it, usually people are held up by matters of identity. They have ideas like, "users don't come from there," or "we're making tools for families, not businesses." If we've learned anything from our controversial modern era, identity is important to people; it's something people can cling to very strongly, even if their identity is preventing other goals. In fact, now that I reflect back on it, the women's fashion app that was able to get tens of thousands of users was run by two men. It's important to note because, they had less identity involved in a women's app than people who are making applications for people like themselves.
I remember once, after looking at the data, I pointed out to my CEO that nearly all our initial sales were $10 items. ($10 was our minimum item price.) After a customer had purchased a cheaper item, they were more likely to want an expensive item for their second purchase. I recommended to my CEO that we start selling more cheaper items as "starter" items that people might buy for their first purchase. He was into it, but some of the high fashion women weren't into that idea because they were worried it would cheapen our brand. Effectively, the CEO didn't have any identity attached to being "high" or "low" fashion so was willing to act on evidence. Yet, people who were more invested in fashion identity were resistant, because they wanted to be a certain type of fashion company.
Similar kinds of resistance have cropped up all over the place in my career, and they're not always without merit. For instance, it might have been a good idea to keep higher status brand, but that needed to be a tactical decision rather than an emotional one. One compelling counter argument was, we were having a few discussions with some celebrities about some sponsorships, and given the power of celebrity sponsorships, it might have been worth our while to cater to that celebrity's tastes (presumably high fashion.) But, it's usually not this type of tactical trade off that sinks startups; it's a kind of unconscious attachment to being a certain type of company without much introspection on if that's serving the purposes of user acquisition.
Because, end of the day, for new projects or new companies you don't have a lot of space for personal considerations. You have to go after sales, hard – it must be your primary goal. (Nearly) everything else can be sacrificed, but users cannot. Without users, you have no company; users trump everything, even tech. And, in my experience, many places don't really get that. I've even heard people be derisive or mocking toward their own users; big mistake. The first step for any successful company is appreciation and understanding of their own user base.