7 ms·
How Segment Found Product Market Fit
- bradgranath 4y agoGoes to show just how much of a scam VC really is.
- echelon 4y agoWorked out for the VCs just fine. And who would realistically try these things without VC? Not many of us are so fortunate to self fund.
- ok_dad 4y ago> Worked out for the VCs just fine. Scams usually work out for the scammer, yes.
- yawnxyz 4y agoWorked out for those who acquired Segment too— they make real money, AND provide a real, very useful service!
- aaur0 4y agoThis post is gold. We are going through our journey of finding PMF. The points mentioned here are what we learned the hard way. Highly recommend it to anyone who wants to build a startup or works for an early-stage company.
- phanis 4y agoYes! I am thinking discovering PMF is the toughest part of a startup journey. Wish you the very best in the journey!
- mittermayr 4y agoI think the hardest part about this (the article is inherently tainted a bit by survivorship bias of course) is to know when exactly it is time to move on from an idea, and when to try a bit harder with it. A fantastic idea posted on the wrong forum and/or at the wrong time feels EXACTLY like you've missed PMF (especially for someone doing this for the first time). I believe there is a lot of history of this right here in the HN archives. Just because nobody picks up on it, doesn't automatically disqualify it. Unless, of course, you're 400k of Google ads deep without much of a sign-up, then you might consider moving on. I'd argue that it's next to impossible to know if an idea has failed reaching PMF, when you haven't solved the "getting it in front of the right people" part yet.
- enos_feedler 4y agoIt isn't always hard to know when to move on. I think a lot of it depends on how big the exploration space is around the product you are trying to build. Building a "I'm confused" feedback button for students attending college lectures doesn't provide much roadway for exploration. You can find out pretty fast whether the product has legs. When the product is something like "a low code developer platform", the space of building is so rich it would be worth building for long stretches of time to learn the right thing to make. The difficult part is in nailing the right product. I would keep working on something like this until someone else has won the market. I am not judging ideas by saying one of the above is better than the other. It's quite possible discovering the lecture solution gets you to a bigger business faster and can create more impact and a bigger company. It just seems like it's easy to weigh the design space of the product, the current state of the market to figure out how long you should work on it.
- dehrmann 4y agoI'll pick on Wordle or Among Us. There was something about pandemic times that made those work, but they could have been done any time in the past 10+ years and gotten less attention. Among Us even predates the pandemic. It was just before its time, but not by too much.
- mittermayr 4y agoGreat example actually. Among Us really picked up once everyone was sat at home and I would even argue many people installed a multi-player game for the first time in their lives with this.
- goopthink 4y agoPeter Reinhardt (Segment’s CEO) goes in depth about this journey (and more) in the “Learning How To Sell” episode of Patrick O’Shaughnessy’s “Invest Like the Best” podcast. Likely the original source for this post: - https://open.spotify.com/episode/61c5XhkcxEave8Cd6j55w1?si=F-b2EEhKQb21ImKKpEFFSw https://open.spotify.com/episode/61c5XhkcxEave8Cd6j55w1?si=F... - https://american-podcasts.com/podcast/invest-like-the-best/peter-reinhardt-learning-how-to-sell-founder-s-fie https://american-podcasts.com/podcast/invest-like-the-best/p...
- remram 4y agoThis seems to be Segment: https://segment.com/ https://segment.com/ I don't know what they do. I couldn't find the open source library in question either.
- DanielVZ 4y ago> Segment collects events from your web & mobile apps and provides a complete data toolkit to every team in your company
- remram 4y agoBut what does that mean? Do they receive event data or timeseries and graph it?
- MAGZine 4y agothey forward those events to all of your other systems (e.g. CRM, webhook, helpdesk). they can also do some data health/quality monitoring.
- remram 4y agoIt seems that this open source library can directly submit web tracking data to multiple platforms, like Google Analytics. If this is the case I don't understand what Segment the company does. I understand that I am probably not their target population, but I really can't decipher the buzzwords on their website.
- joegahona 4y agoNo. There is no data visualizer inside of Segment. It’s a very expensive customer-data-infrastructure tool.
- gmt2027 4y agoIntegrate segment with your application on the web and mobile and you can enable or disable downstream data delivery to any one of the services at https://segment.com/catalog/ https://segment.com/catalog/ without touching your code. This leaves your engineering team free of requests to integrate your product with new services your marketing and analytics teams want to use.
- Jasper_ 4y agoThe posts like this are all fantastic, because they're all "we built a dumb product in two weeks that nobody wanted, and then when that didn't work, we built a data harvesting rig and everybody threw money at us". It's just the Mitch Hedberg joke about "that inspired me to make a movie about a gorilla" My other favorite was the one that started with a dating app, then a corporate benefits HR package solution, before eventually finding "product market fit" and "success" by turning it into an internet payments company, with the founder going on all the podcasts to talk about all the lessons they learned. Then they shut down a year later, with the founder getting a hefty parachute for their wonderful invasive innovations they generously thrust into our world.
- ramraj07 4y agoIsn’t this the story of every post 2000 tech company? The prime examples being Google and Meta. In some generalizational way isn’t this the story of all capitalism? “We built the best baby formula ever, but then we realized our customer isn’t the mother, it’s the stockholder so we started caring more about making money product quality be damned.”
- khazhoux 4y ago> Isn’t this the story of every post 2000 tech company? The prime examples being Google and Meta. Not really... Google: "We built a search engine, and everyone who saw it immediately loved it. Then we sold ads and got rich." Meta: "We built a social website, and everyone who saw it immediately loved it. Then we sold ads and got rich."
- SQueeeeeL 4y agoNot enough acquiring random rival competition to prevent them from competing and optimizing engagement so hard you get a game show host elected president
- allenu 4y ago> This case study makes me wonder if PMF is only accidental. Many startups I know have accidentally stumbled on the idea that made them big. This is one such story. It does seem accidental to me in this particular case. The more I read about successful product-market fit, the more I'm convinced that even if it's a real thing, knowing about it doesn't actually lead to any useful advice other than "keep pivoting until you've made something people really want" (or else you run out of money). The term "product-market fit" makes it seem like a concept that is measurable. Maybe, it's measurable, but only after you've achieved it. So what does that tell you exactly? If you're not successful, you don't have it. If you are successful, you must have had product market fit. Amazing. Now go forth and make stuff that has product-market fit. Good luck. It feels more honest to just just call it "making a product people desperately want and will you give you money for". It would simplify a lot of discussions. "We worked on this product and didn't know if people desperately wanted it or would give us money for it. Then we released it and realized people didn't want it and would not give us money for it. The lesson we learned is we should make something people desperately want and will give us money for." Apologies for being negative, but I think I've watched too many videos and read too many articles from successful people/companies and am starting to think that survivorship bias and luck gives too much authority to advice that is actually not that useful.
- drc500free 4y ago> "making a product people desperately want and will you give you money for" That’s all pmf has ever meant though, right? It’s just fewer words to have a label for the concept.
- samcal 4y agoYeah, it's even in the OP: "Users saying that it's a good idea / product after a demo doesn't mean that the product has PMF. Users desperately wanting to use the product and pay for it ie., retention / repetitive usage is PMF." The bigger issue is clearly that the advice boils down to something more like "if you don't have buyers, you won't make sales", which is just common sense and I'm surprised we need to have a whole buzzword about it.
- 4y ago
- ilrwbwrkhv 4y agoI think for a lot of founders transitioning over from being developers (I followed this path), it's important to create a simple product which is free and gets at least 1000 uses. This, I feel, is a missing critical step. To create something which can stand on its own, and is actually useful, is much harder especially when you have been working for a company for sometime, where a lot of things that you do don't actually matter (not because you are bad, it is just the nature of companies). A lot of startup reading material talks about immediately charging for your product and while I agree charging is the only proof of a viable business, it is in my opinion step number 2. Step 1 is building something which people use repeatedly and this product instinct is almost like a muscle that needs to be trained. Once you create something which gets used a 1000 times (either by 1000 different people or a smaller number of people using it multiple times), you get a taste for it and almost a subconscious feel for what works and what won't.
- afhammad 4y agoIts difficult to say which is the correct approach, but one thing I know is pricing is very hard and getting it wrong is costly. Getting usage first allows for constant valuable feedback but also establishes a value on the product within the user’s mind. What is this worth to them? How much time is it saving them?
- auggierose 4y agoSo how would you classify what Apple did with the iPod and then the iPhone? Did it define and shape the future? Or did they find out what the market wanted and solve for it? It seems to me both approaches are valid. One approach builds the future that most people imagine and want. The other one builds a future that one person, or a small group of people, imagine and want. Certainly the former is more likely to succeed, but the latter might yield more interesting results if successful.
- harrisonjackson 4y agoApple also wasn’t anything close to a startup when they released the iPod or iPhone.
- rsanaie 4y agoAlso not all founders are visionary geniuses like Steve Jobs, with confidence the size of Eiffel tower
- auggierose 4y agoI'd rather have visionary geniuses as founders, though ...
- yawnxyz 4y agoPieter Levels says that he's tried something like 70+ ideas to make money and only 3 of them have really made some money, and only one have done it substantially. So his conclusion is that you need to get good at trying new things / shoot goals, because it DOES come down to "I have no idea what will stick". But the more shots you shoot, the more likely. But also— most of us are in software, and every product we build (including Segment) has compounding effects towards the next thing we build. The frameworks we use or build can inform our next thing. The account flows. The databases and CI/CD we get used to. For every thing we try, we get a little better at the next thing. I've found that "launching ideas" from a napkin sketch to working prototype has become significantly easier. Used to take me a month+ to launch a crud idea/prototype; now it takes me a weekend. So the limiting factor for me isn't building/launching anymore... it's customer discovery and sales.
- woutr_be 4y agoI truly wish I could launch something in a weekend. In retrospect, many of my ideas should've been just weekend projects, but I somehow manage to spend months on them, only to have second thoughts and throw them away.
- ttul 4y agoThis story reminds me that the most successful entrepreneurs are often those who have wealth to begin with, which enables them to keep playing with ideas until the random walk bumps into something that customers actually want. Segment’s founders may not have raised another $600K had their search ultimately failed. And did the founders have the cash themselves to try again? The panic attacks indicate perhaps not.
- spaceman_2020 4y agoAs a bootstrapped founder, my execution has been markedly different after I had a small exit. I started with enough cash in the bank to pay for 3 months of operations. Now, I have enough for 30 months of ops. It's a massive change in the way I run things now. Money is the single most important factor in anything, from creating art to creating companies.
- dna_polymerase 4y ago> It's important to shut down bad ideas. Codecademy Founders tried 12 ideas in 7 weeks. Just 4 days before the YC demo day they came up with the idea of Codecademy, which had a PMF! That strikes me as such an odd way to found a company. Founding, only for the sake of founding. I find it hard to believe that companies founded like this will ever succeed. (at least in the average case)
- allenleee 4y agoTL;DR - Product-market fit is a spectrum, as opposed to something you definitely have or don’t - Every business is unique, and metrics frameworks apply differently. Focus on the metrics that matter for your product, and make sure they’re clearly defined - Benchmark: Users scream for the product so you can screw everything up and still win (for a period of time) - Prove a value hypothesis and only once you've proven the value hypothesis should you test a growth hypothesis. PMF is when you have proven the value hypothesis. Product Market Fit (PMF) can rise and fall as the product and market change and grow. It’s not this static thing that once you get it, you always have it. I found Andy Rachleff's thesis on PMF to be the most realistic: *B2B* Every enterprise company should run a 30-day proof of concept trial. To make these trials productive, company usually requires the customer to agree to buy the product after 30 days if it meets expectations. But… pull the trial after 30 days (no matter what). Then: “If the customer doesn’t scream, you don’t have product-market fit because if they’re not going to buy it at the end of 30 days, they’re not desperate, and if they’re not desperate, you don’t have product-market fit.” “The second biggest mistake I see entrepreneurs make, especially in enterprise, is when they pitch a potential customer on an idea, and when the customer doesn’t like the idea, they try to iterate on the product to build something the customer would want.” “That’s the absolutely wrong thing to do, even though it feels right. You want to find people who love what you’re doing, not try to convince the ‘no’s’ and turn them into ‘yes’s.'” *B2C* “The only way you know if you have product-market fit is if you get word of mouth” (and the best test of word of mouth is exponential, organic growth)
- teddyh 4y agoApparently user fingerprinting is where the money is. (This has been a re-post of this close to three year-old comment about another hype-up about Segment: https://news.ycombinator.com/item?id=21709363 https://news.ycombinator.com/item?id=21709363)