Do you agree with Linear that it's time to rethink the MVP?
Chris Messina
29 replies
I agree that the MVP is a journey, and not a one-time validation event, and that continuous iteration is necessary to refine the product into a competitive offering in existing markets.
I found the emphasis on using a waitlist strategically to collect targeted feedback from specific tranches of early adopters a solid recommendation.
Given this, it tracks that when you narrow your target audience, you can be more selective and intentional on how you deploy resources. By leveraging deep understanding of specific user needs, you can create significant value compared with the competition.
Is this similar or different to your approach?
Replies
Justin Mitchell@jmitch
Cleft Notes
I see it more as a combo. You start with minimum features to validate. Once that’s done you validate again and again through multiple iterations of mvps.
Share
Mammoth 2
Cleft Notes
@chrismessina it may but depending on the market that’s not exactly a bad thing. Starting as narrow as you possibly can with your product and audience is key to success.
Agree Chris
Hey Chris, good to see #you!
There is LOTS to agree with in that blog post by the Linear folks. Kudos to that team for leading us to better ways than flogging JIRA over and over and over. My thoughts here are not a take-down of the Linear team.
BUT. Two things:
ONE:
Let’s zero in on the mis-use and now mis-expansion of the MVP concept. Eric Reis (the originator of the term) used “Minimum Viable Product” to refer to TESTS. Ways to validate the value of a POTENTIAL product before building the actual product. In Eric’s world (which is the same as the world of MVPs), one would never actually charge for use of an MVP experiment.
For many reasons, I think we should stop using the term MVP. It is overloaded and continually misunderstood, and therefore does not create common ground to walk when used in conversation and analysis.
TWO:
Missing the spirit of Eric’s original MVP concept, the Linear folks skip an important and high-value step in the description of their newly proposed multi-step MVP process.
Their process begins:
1. You start with the idea on how you want to improve on existing products in a category.
2. You build your first prototype.
3. You iterate with your vision and based on feedback from early users.
OOPS! Missing the ethos of the original MVP concept, that process rushes to building a prototype.
INSTEAD, invert items two and three. A MUCH MORE VALUABLE AND LOWER COST version of the Linear.app process would be:
1. You start with the idea on how you want to improve on existing products in a category.
2. You iterate with your vision and based on feedback from early users.
2.1 You run the interview/iteration process multiple times, zeroing in on concrete value to deliver. Once you are confident enough in your solution description:
3. You build your first prototype and socialize it to your early feedback audiences.
4. You get an inkling of product-market fit.
5. You repeat the process, refining your low cost prototype and acting on market feedback multiple times.
6. After alpha-like and beta-like releases, you ship your GA release with already-collected success stories from the alpha and beta phases.
And so on…
CONCLUSION:
ONE. Ban the term MVP.
TWO. Use prospect and customer conversation to refine your potential solution multiple times before even starting to commit code.
Saying “MVP is a journey” feels a bit trite to me. All product work is a journey, what we call the milestones along the way to PMF and onward can be fairly arbitrary, and that’s okay in my book. Emphasizing an “MVP” release may be myopic, but it is a reasonable tool in the toolbox for teams that regularly over scope.
Waitlists work well if you, and your target audience, are patient and deliberate. I think they tend to backfire for teams who lack confidence. For those teams, I’d suggest releasing product to market early and often. Throw them into the fire, so they can get over their fear of negative feedback and learn to treat it all like a learning opportunity.
On being highly selective with a target audience, everything you wrote is true. The one caveat is the risk of bias in the selection criteria. It’s very easy to accidentally create an echo chamber and miss critical insights from all of the users you don’t have.
One stage boatloads the next. That's how I think of it. At each stage you should create enough value to get to the next stage.
This sounds similar to building off the market. In that case, developing hypotheses and identifying a narrow niche of customers or users' pain points of similar products or specific customer behavior makes sense. Effectively, the product is built on customer feedback.
It took two and a half years to get the Airtable out of stealth, and it was invite-only back then with a focus on "crowd-sourcing" unique ways to use the platform and encouraging creativity. And Notion started as a community on Reddit.
These companies sell a lot of B2B, but B2C acquisition worked as a channel to bigger sales. It is the same as Slack's/Dropbox's critical number of team members of a company/team when the enterprise purchases the software. However, not all companies are in the same category of product's ability for PLG.
Using this or a different approach depends on a given startup's unique situation and environment: a combination of a team, idea, resources, and luck.
Space Agent
An MVP is aimed at proving value. Why delay receiving information that would accelerate providing value and validating a better direction?
Crafting a successful product is akin to embarking on a journey marked by milestones. Most milestones, particularly when launching a whole new feature set, can be seen as a series of MVPs. Treat them as such and the MVP becomes a distilled version of a product or feature set, crafted to deliver core value while mitigating risks to the overall product and brand. Once launched, the iterative process begins: gathering feedback, refining features, and enhancing user experience.
I think MVP is only a framework to create the right level of focus and feedback. If a team has other ways to focus on what’s fundamentally important, and to start gathering feedback sooner rather than later, then they would probably get comparable results to a team that works on the MVP framework.
Totally agree. It's easier to tell the story of that target user's journey and also say "no / not yet" to features or unrelated journeys.
Kraftful
Completely agree that MVP is a journey. But not sure that waitlist is a good idea for all situations. If you’re in a crowded space that moves quickly, you’re better off launching early (a la Y Combinator) to generate traction and potentially get first mover advantage. You just have to be careful to still focus on a narrow subset of users when refining your MVP.
I think there’s a lot of fear that people will see a half-baked product and never come back. But in reality, no one remembers. Users will use whatever solves their problem. If they try out something that didn’t solve their problem originally, they will try it again in a year if everyone is raving about it then. I guess this too is YC wisdom: just relaunch until someone notices.
Product Hunt
@yanaw I agree. Whats the upside of waitlist - No bad reviews or ratings? You get it "right" before your actual release? Create FOMO? These are all valid but they also create friction and everything is moving so fast, tomorrow is a new day. No one is NOT going to use the next iteration of Gemini for example.
Kraftful
@thomas_benham2 Exactly! I keep trying every single one of their models hoping that the next one will be better 😅
Kraftful
@thomas_benham2 Yeah, I was just thinking that they certainly should have launched (not marketing launched!) sooner to avoid setting high expectations.
You haven't missed much! Here's a side-by-side that I did with one of Kraftful's prompts to extract feature requests from Nest app reviews. At full context window, Gemini Pro didn't follow simple instructions to output ONE list, let alone summarize the feature requests: https://x.com/yanatweets/status/...
I've tried the later models too. Still not impressed.
@rajiv_ayyangar, feature request for your team: I would have loved to attach an image to this post instead of linking to Twitter 👍
I think so. Taking a product takes time to build tbh. Even today. But getting a Carrd or Framer one-pager up and running even with a domain redirect to gather waitlist emails and then ask them questions either via email or form is better than just not having that at all. Waitlists are a good way solve for the cold start problem. The MVP or Alpha version of a product comes after... it's what we're seeing on https://early.tools – it's more of a linear (no pun intended) progression of graduating from one phase to the next of building. Waitlists are the easiest way to get started whilst you build a product for weeks or months if it takes longer.
Really recommend looking at https://early.tools (which I am building). So many early products on there that are using waitlists. Just makes sense tbh @chrismessina!
Agree the MVP approach helps to fast track the creation process of a new product without market fit well defined.
That's allow to quickly validate hypothesis and knowledge gap instead of spending months to build advanced product without market adoption validation.
Into big product MVP are also used as Beta to apply this approach to new features launches.
Wallafan
It's very late here at the time I'm writing this, so forgive me all if I'm misunderstanding the topic here. To me an MVP isn't different from the whole picture.
The project that is sucking my life out of me right now is a SaaS. We have started with an MVP and bootstrapped the first public release one month ago, and our (mine and team) approach has remained the same then since and for the future: forget short term profitability and focus on long terms value.
I have written an article on Linkedin about it - https://www.linkedin.com/pulse/d... - and although not directly related with rethinking MVPs, I think it can still...relate.
MVP is just a label for a milestone in the life of a project/product, but like every life's milestone it defines where you want to go. I've never understood businesses that refuse to release in-house much needed features at an early stage due to product-maturation marketing tactics or those who charge for the essential features as if they were upgrades.
You can't sell me a toilet and then charging me extra for the running water or telling me that it will come at a later time (sorry for the gross example): a minimum viable toilet MUST come with running water, or it is not viable to me, no matter how cute, good-smelling and futuristic might be.
I tend to agree with using a waitlist and gathering a good understanding of specific user needs, but from different audience segments. Here's my reasoning.
If you are only focusing on one specific target audience, you might be missing out on an oportunity to branch out into other market segments and break away from the competition.
There are four types of customers :
1. Current customers - people who are already in your specific market. To get their attention and win over the competition you've got to offer them an exceptional value, so they make a switch to your solution. You will always be compared to others. For example, someone using a pencil is looking for one that lasts longer, is less prone to breaking, is sharper, cheaper etc. It's a feature comparison between other pencils in the market so they will only compare and give you feedback on the things you and they already know. It's super competitive and all you're trying to do is be slightly better than your competition.
2. Non-customers - who are at the edge of your market and ready to become customers. These people are evaluating new solutions to their problems and looking for ways to improve their processes. To get their attention you've got to show them how much better their life is going to be if they use your product. To use the same example, people who use a pencil to write are faced with a problem of it breaking, so they've got to sharpen it and it wastes a lot of their time. You just came up with this amazing idea of a pen. It lasts longer, doesn't need sharpening...problem solved. They now purchase pens instead of pencils. Now you get feedback based on the comparison of using a pencil vs pen as opposed to using pencil A vs pencil B.
3. Non-customers who are aware of your market's offerings but refuse to join it. A good example would be of people who tried using a pen, but it leaked everywhere and ruined the page they were writing on. On top of that a pen didn't offer the ability to erase and correct mistakes. So they kept using pencils instead. Now imagine you came up with a pen which doesn't leak and you can erase it! You just won over a new customer cohort who had previously refused to join your market. They will give you feedback based on using a pencil, a regular pen and your super pen!
4. Non-customers who are totally unaware as they are in the distant markets. This cohort of people never used a pencil or a pen. Never concidered either of them, because they don't send out hand writen letters. Maybe they make phone calls instead. Now you've a brand new untapped audience to get feedback from, learn about their problems and educate them on the benefits and value of your new alternative solution. You are creating demand!
So yes, going after one specific target audience is easier and you get to build out your product based on very specific needs and requests. But making presumptions on who your target audience is closes the door to much greater potential audiences with less competition.
I'm on the same page about MVP being a continuous journey. It's not just a one-time thing – you have to keep refining your product to stay ahead.
What really caught my eye in the article was the idea of strategically using a waitlist to get feedback from specific groups of early users. It makes total sense; when you focus on a smaller audience, you can be more thoughtful with your resources. Understanding your users deeply gives you a competitive edge.
At JetSoftPro, where we've been immersed in the world of software development for two decades, working closely with startups, we've compiled some invaluable insights about MVP in one of our articles. If you're navigating the startup landscape or interested in optimizing your Minimum Viable Product approach, I highly recommend giving it a read.
https://jetsoftpro.com/blog/supe...
Thinkbuddy AI
In my opinion, rather than filtering the waitlist only for our target audience, it would be more beneficial to categorize all users into different groups. Feedback from your primary target users is of course important, but insights from users outside your target demographic can also provide valuable information. The key point is to divide users into categories and then analyze the feedback from each group according to their specific characteristics. @chrismessina
Mammoth 2
@sezerufukyavuz what tool would you recommend to use to do that?
Thinkbuddy AI
@chrismessina A product like this exists it seems they've devised a good solution for issues of this kind. https://getlaunchlist.com However, people still haven't fully grasped the value of analysis. If there's greater demand for such matters, we might see an increase in the quality of products available as a result.