3 ms·

Advice for first time founders

53 points by edmack 12y ago

confiscate 12y ago
"Technology is bad. It distances you from users. It slows down iterations. It’s expensive to build. It provides a worse experience than you could provide in person or over email." I don't quite agree with the saying that "technology is bad". Obviously like anything else, if you mis-use it, it will not generate the benefits you expect, but unilaterally claiming that "tech is something to avoid as much as possible" seems like pretty bad advice. That's kind of like a pure tech company claiming that "accounting is bad" and "try to avoid accounting as much as possible". Every company needs good accounting at some point. It's foolish to unilaterally assume it will be bad to have it. Good luck finding good engineers to work for your company. Or maybe, good engineers are really not that important for the company / product you're running / building, which is actually fine for a non-tech-oriented product/company. Most companies aren't tech companies and don't need strong engineering, or even any engineering at all in most cases. Just don't call yourself a tech company, when the focus isn't on tech.
- richardbrevig 12y ago
  You have a valid view. I interpreted the author to be re-stating the "do things that don't scale" advice from Paul Graham.
- yeukhon 12y ago
  I think, I think I have a different interpretation. Rule #1: as founder your job is growth. No doubt. If you go solo sure you will probably spend a lot of time coding, but don't forget about growth. I would imagine myself doing 30% of coding, 40% of growth, and 30% of rest (at least 7 hours of sleep and in-between of 40% of growth and coding I probably get to eat lunch, talk to people, watch a couple gameplay on YouTube). That probably would be my life for the first year of my start-up. Rule #2: for many apps, believe it or not, you can engage with your users directly. If you are a mobile app for restaurant, go out there and talk to people about your app. Hey, why not go on the subway and do a little pitch sale? If you are an account SaaS, talk to accountants out there? Social media advertisement is great, and you should do that, but don't forget to engage users without the technology. Your friends are your number one targets. If they can't engage with your product, why would Bob or Alice? Rule #3: When he said avoid building, he's right. New features are absolutely great. But focus on usability and basic feature. Pick either Instagram or Vine. They are great mobile apps. They started out with simple UI and a handful of functions. The early employees probably spent their time trying to meet growth demand rather than feature demand. If your infrastructure is weak, focus on that. Who knows if your app would become viral tomorrow or not? Harden your vms before too late. Don't build an image matching algorithm before you finish your basic search feature... Rule #4: If you are into reporting, don't go crazy on building fancy tool. Add a simple contact us / feedback button with a dialog. Don't build your fancy dashboard. That's for later. If you know ELK and you have done configuration management like I do as a daily job, sure, you can use ELK to do neat reporting instead of using GA (there is a whole GA discussion on HN tonight). But I will probably leave building ELK for a weekend...or do span that across multiple weekend. Have you done proper logging in your code yet? No? Go back and add some more logging statements. Building a startup is exactly like doing iterative software development. Plan something, execute, and reflect. Deliver priority, shovel other things into backlog. If you have co-founders or you already have hired early employees, you definitely need to do daily stand-up, and periodically tell the world about the company's progress! Ship basic features. Improve performance, resiliency, security and usability. Sell your product to users. Write blog posts, tutorials, videos so users know what the start-up is up to. You are a human. This is how I am going to build my start-up. The experience counts.
brandonb 12y ago
There are many truths here, but I keep seeing HN articles with variants on the following piece of advice: "As a founder your sole aim is growth. If you don’t have growth now, you are failing." That's only half-true. To simplify a bit, there are two types of startups in the world: Twitter, and the hoverboard. Twitter is easy to build, but the question is: does somebody actually want it? Twitter startups are all about market risk. Conversely, if you could build a hoverboard, then of course people would buy it. Hoverboards are awesome. The question is, can you actually build it? Hoverboard startups are all about technical risk. The advice to focus on nothing but growth applies to startups where market risk is the dominant risk. You need to validate that there's actual demand for your product. And that's where advice like "If you don’t have scale, you probably don’t need (much) technology", "try no-tech," "[edit] static content and embedding some forms" applies. But if you're building Tesla, or a cure for cancer, or self-driving cars, or most enterprise products, or any number of startups which require a technical breakthrough, you literally cannot focus on growth from the start. The minimum viable product for these types of areas may take months or even years to build. The experiments you should run for those types of startups should validate technical assumptions: for example, can we make a battery energy-dense enough to power an electric car? Or they should validate market demand through customer development, which is actually quite useful in many areas such as enterprise sales where achieving growth may take 6-18 months even in the best case. In general, when you hear startup advice, it's wise to remember whether the author's startup comes from the world of technical risk or market risk. The right things to do for each type of startup are often complete opposites, and it's easy to get confused since most people give advice based on what worked for them, but don't necessarily include the context which circumscribes where the advice applies. As always, Steve Blank said it first and better: http://steveblank.com/2009/05/28/vertical-markets-2-customermarket-risk-versus-invention-risk/ http://steveblank.com/2009/05/28/vertical-markets-2-customer...
- timr 12y ago
  "The advice to focus on nothing but growth applies to startups where market risk is the dominant risk." Yep. But there's also a class of startup that we might call the Secret Hoverboard: it's the sort of problem that looks like a Twitter ("would people want to get more stuff delivered if it only costs $5?"), and when you poke at it with the stick of "doing things that don't scale", you see strong consumer demand -- market validation! Secretly, however, there's a hoverboard-sized technical problem between manual labor and profitability (e.g. "if we can just get our fleet of bicycle messengers operating with X% efficiency we can deliver anything by bicycle for a flat $5 surcharge, which will undercut the business models of all existing delivery services!") These are tricky, because people are so used to market risk being the dominant factor that they can be fooled into believing that the strong consumer demand means that the rest is easy. But really, it's telling you that people love to get something for nothing, and the market got the solution right the first time.
zkhalique 12y ago
Hmm, we definitely have a different view. http://qbix.com/blog/index.php/2011/04/business-models/ http://qbix.com/blog/index.php/2011/04/business-models/ 1. People live lives. Companies create products. 2. Make a business model that works 3. Make it safe to fail over and over, so you can iterate 4. Measure everything 5. Nail your engagement metrics 6. Take advantage of viral loops and optimize them 7. Optimize engagement 8. Optimize monetization 9. Prepare onboarding process and system for future hires, including Partnering Per Project agreements 10. Attract great people by following 3 - 7 11. Reinvest into the company. Optional: 12. Raise money from investors following 3 - 7 Repeat endlessly until mission accomplished.
dmritard96 12y ago
Many points DNA to hardware/software startups fyi.

53 points by edmack 12y ago

"Technology is bad. It distances you from users. It slows down iterations. It’s expensive to build. It provides a worse experience than you could provide in person or over email." I don't quite agree with the saying that "technology is bad". Obviously like anything else, if you mis-use it, it will not generate the benefits you expect, but unilaterally claiming that "tech is something to avoid as much as possible" seems like pretty bad advice. That's kind of like a pure tech company claiming that "accounting is bad" and "try to avoid accounting as much as possible". Every company needs good accounting at some point. It's foolish to unilaterally assume it will be bad to have it. Good luck finding good engineers to work for your company. Or maybe, good engineers are really not that important for the company / product you're running / building, which is actually fine for a non-tech-oriented product/company. Most companies aren't tech companies and don't need strong engineering, or even any engineering at all in most cases. Just don't call yourself a tech company, when the focus isn't on tech.

richardbrevig 12y ago
You have a valid view. I interpreted the author to be re-stating the "do things that don't scale" advice from Paul Graham.
yeukhon 12y ago
I think, I think I have a different interpretation. Rule #1: as founder your job is growth. No doubt. If you go solo sure you will probably spend a lot of time coding, but don't forget about growth. I would imagine myself doing 30% of coding, 40% of growth, and 30% of rest (at least 7 hours of sleep and in-between of 40% of growth and coding I probably get to eat lunch, talk to people, watch a couple gameplay on YouTube). That probably would be my life for the first year of my start-up. Rule #2: for many apps, believe it or not, you can engage with your users directly. If you are a mobile app for restaurant, go out there and talk to people about your app. Hey, why not go on the subway and do a little pitch sale? If you are an account SaaS, talk to accountants out there? Social media advertisement is great, and you should do that, but don't forget to engage users without the technology. Your friends are your number one targets. If they can't engage with your product, why would Bob or Alice? Rule #3: When he said avoid building, he's right. New features are absolutely great. But focus on usability and basic feature. Pick either Instagram or Vine. They are great mobile apps. They started out with simple UI and a handful of functions. The early employees probably spent their time trying to meet growth demand rather than feature demand. If your infrastructure is weak, focus on that. Who knows if your app would become viral tomorrow or not? Harden your vms before too late. Don't build an image matching algorithm before you finish your basic search feature... Rule #4: If you are into reporting, don't go crazy on building fancy tool. Add a simple contact us / feedback button with a dialog. Don't build your fancy dashboard. That's for later. If you know ELK and you have done configuration management like I do as a daily job, sure, you can use ELK to do neat reporting instead of using GA (there is a whole GA discussion on HN tonight). But I will probably leave building ELK for a weekend...or do span that across multiple weekend. Have you done proper logging in your code yet? No? Go back and add some more logging statements. Building a startup is exactly like doing iterative software development. Plan something, execute, and reflect. Deliver priority, shovel other things into backlog. If you have co-founders or you already have hired early employees, you definitely need to do daily stand-up, and periodically tell the world about the company's progress! Ship basic features. Improve performance, resiliency, security and usability. Sell your product to users. Write blog posts, tutorials, videos so users know what the start-up is up to. You are a human. This is how I am going to build my start-up. The experience counts.

brandonb 12y ago

There are many truths here, but I keep seeing HN articles with variants on the following piece of advice: "As a founder your sole aim is growth. If you don’t have growth now, you are failing." That's only half-true. To simplify a bit, there are two types of startups in the world: Twitter, and the hoverboard. Twitter is easy to build, but the question is: does somebody actually want it? Twitter startups are all about market risk. Conversely, if you could build a hoverboard, then of course people would buy it. Hoverboards are awesome. The question is, can you actually build it? Hoverboard startups are all about technical risk. The advice to focus on nothing but growth applies to startups where market risk is the dominant risk. You need to validate that there's actual demand for your product. And that's where advice like "If you don’t have scale, you probably don’t need (much) technology", "try no-tech," "[edit] static content and embedding some forms" applies. But if you're building Tesla, or a cure for cancer, or self-driving cars, or most enterprise products, or any number of startups which require a technical breakthrough, you literally cannot focus on growth from the start. The minimum viable product for these types of areas may take months or even years to build. The experiments you should run for those types of startups should validate technical assumptions: for example, can we make a battery energy-dense enough to power an electric car? Or they should validate market demand through customer development, which is actually quite useful in many areas such as enterprise sales where achieving growth may take 6-18 months even in the best case. In general, when you hear startup advice, it's wise to remember whether the author's startup comes from the world of technical risk or market risk. The right things to do for each type of startup are often complete opposites, and it's easy to get confused since most people give advice based on what worked for them, but don't necessarily include the context which circumscribes where the advice applies. As always, Steve Blank said it first and better: http://steveblank.com/2009/05/28/vertical-markets-2-customermarket-risk-versus-invention-risk/ http://steveblank.com/2009/05/28/vertical-markets-2-customer...

timr 12y ago
"The advice to focus on nothing but growth applies to startups where market risk is the dominant risk." Yep. But there's also a class of startup that we might call the Secret Hoverboard: it's the sort of problem that looks like a Twitter ("would people want to get more stuff delivered if it only costs $5?"), and when you poke at it with the stick of "doing things that don't scale", you see strong consumer demand -- market validation! Secretly, however, there's a hoverboard-sized technical problem between manual labor and profitability (e.g. "if we can just get our fleet of bicycle messengers operating with X% efficiency we can deliver anything by bicycle for a flat $5 surcharge, which will undercut the business models of all existing delivery services!") These are tricky, because people are so used to market risk being the dominant factor that they can be fooled into believing that the strong consumer demand means that the rest is easy. But really, it's telling you that people love to get something for nothing, and the market got the solution right the first time.

zkhalique 12y ago

Hmm, we definitely have a different view. http://qbix.com/blog/index.php/2011/04/business-models/ http://qbix.com/blog/index.php/2011/04/business-models/ 1. People live lives. Companies create products. 2. Make a business model that works 3. Make it safe to fail over and over, so you can iterate 4. Measure everything 5. Nail your engagement metrics 6. Take advantage of viral loops and optimize them 7. Optimize engagement 8. Optimize monetization 9. Prepare onboarding process and system for future hires, including Partnering Per Project agreements 10. Attract great people by following 3 - 7 11. Reinvest into the company. Optional: 12. Raise money from investors following 3 - 7 Repeat endlessly until mission accomplished.

dmritard96 12y ago

Many points DNA to hardware/software startups fyi.