7 ms·

Why Founders Fail: The Product CEO Paradox (2013)

173 points by ggonweb 11y ago

salimmadjd 11y ago
OT: There is a bug in this blog that when I tried to share it on Facebook, it shows: "<a data-width="300" data-height="200" data-bop-link..."
- scintill76 11y ago
  I saw a shared post on Facebook the other day with the title "302 Found". I'm guessing it was taken from the standard generated HTML Apache etc. will send when redirecting. Not sure if it's more Facebook's or that site's fault.
ryandrake 11y ago
Great article. While it's great to have a CEO who stays engaged with the product even as the company scales up, it's awful when they continue to micro-manage it the way they did when the company was 10 people. You hired product managers (and perhaps product manager managers). So let them do their jobs and create great products! When you're running a 500+ person company, you're not going to go into the code and start optimizing the graph traversal algorithm--you hired smart engineers to do this. You're not going to fire up Illustrator and start drawing buttons--you have artists for this. You're not going to write the company blog yourself, or fix the automated test systems yourself, or design the marketing materials yourself. So why do you feel the need to over-ride your product managers on product direction, feature set, user acquisition plans, retention strategy, etc? Doing so de-motivates the professionals you have whose job is to do these awesomely.
- mpdehaan2 11y ago
  Product managers are potentially both the most helpful and potentially most damaging hire you can make, as you are ceding them control (if you do) over the very definition of the value your company provides. I think someone like Steve Jobs keeping a lot of that control there is definitely of value - in not delegating it. I've worked with a lot of PMs that were like "this dialog needs to be more orange" (one defunct startup from many years ago), and miss the picture, because they are just kind of faking it -- they are the guy with the opinion and the title, because somebody had to do it, but not really trying to make the best choice for the end user. There are some that could be exceptional. (Speaking of several different companies where I've experienced this). You need to have someone with a really good gut feel and instinct, and a lot of passion to believe in the thing they are doing. But if a PM is a "CEO of a product", that makes it the most important choice in your entire organization -- for that product. If you have one product, that's huge. An alternative approach is to get a company really involved, to where everybody on the team has input in the product, and that you source the ideas from everywhere, but yeah, somebody still has to drive it. Depending on who you talk to PM can mean like 37 different things, and no one person can concentrate on all 37 -- you do need people doing all of those things, but I think it's GREAT if the person at the helm of your company is all about the product. Else, you're just cranking size and numbers? The purpose of the company, to me, is about creating the things you believe in and want to create - the business is the means to the end, and the food on the table along the way. Approaching it from a pure business perspective is missing a lot of the fun and opportunity.
  - omouse 11y ago
    >Product managers are potentially both the most helpful and potentially most damaging hire you can make, as you are ceding them control (if you do) over the very definition of the value your company provides. I think it ties into what paul graham wrote about hackers making things for users. The more layers you have between the hackers and the users, there's more potentially that your product won't be that great. I've worked with product managers who supposedly talk to the users directly and channel that into the work us developers have to do but they always seemed to just copy the competitor or only make small changes whose value is questionable. >An alternative approach is to get a company really involved, to where everybody on the team has input in the product, and that you source the ideas from everywhere, but yeah, somebody still has to drive it. It's probably better to create the position of product manager from within.
- teacup50 11y ago
  Unified vision matters. Peer commentator already noted Steve Jobs, but I wanted to focus on this. Steve didn't draw the buttons, design the hardware, or write the code -- but he sure as shit told the relevant teams exactly what he wanted -- often directly, with a very shallow management hierarchy -- and told them to try again if they didn't pull it off. At the same time, he did have SVP-level and VP-level people writing serious, core OS-level code themselves. They were better managers by virtue of actually understanding and having a coherent vision for what it was they were managing. If the management chain doesn't set product and marketing direction at a company whose purpose is to sell products, then what the hell are they doing in charge?
  - teej 11y ago
    "At the same time, he did have SVP-level and VP-level people writing serious, core OS-level code themselves." I find this fascinating. How can I find out more? Which VPs? Which code?
    - teacup50 11y ago
      https://en.wikipedia.org/wiki/Bertrand_Serlet https://en.wikipedia.org/wiki/Bertrand_Serlet https://en.wikipedia.org/wiki/Avie_Tevanian https://en.wikipedia.org/wiki/Avie_Tevanian Bertrand wrote malloc, top, and quite a bit of code at NeXT. Avie wrote Mac Missles! in the 80s: https://www.youtube.com/watch?v=Ir8H0NuPZRU https://www.youtube.com/watch?v=Ir8H0NuPZRU ... and, well, Mach: http://www.cs.ubc.ca/~norm/508/2009W1/mach_usenix86.pdf ftp://ftp.cs.cmu.edu/project/mach/doc/unpublished/exception.ps ... etc etc. Steve Jobs relied on extremely technically capable management that didn't just hand off understanding to their staff.
- slyall 11y ago
  I had a CEO like that once. He was very good at giving them impression that he could do the job of anyone in the company better than that person and the only reason anyone else was hired was cause he couldn't be in 200 places at once.
  - noskynethere 11y ago
    Can you elaborate a bit more? Could he not do other people's jobs, not do them as well? If he was correct in his assessment, was the issue that it affected morale? Right now, I feel I could do maybe 15 peoples jobs. Not as well, certainly unless I were full time, and so I cede control to them as long as they can explain their decisions. But often I feel like that statement, even if true, bodes poorly for the company. Is that because of how employees respond to that attitude?
    - slyall 11y ago
      He certainly couldn't do many of the technical guy's jobs as well as they could. Even if he'd run the whole thing by himself 5 years earlier he had not kept up with the new systems when everything had grown by 10x. Even for relatively simply things like helpdesk jobs he could just "make it so" since he was the CEO (eg just give customer a credit) rather than having to go though the normal procedures but his technical knowledge was still out-of-date there. He wasn't the nicest person in general. The problem was that he assumed he could jump into any situation and was as expert as any other staff member. So he'd start telling the network guys what to do to solve something (and trying to login to the routers) despite having virtually no idea how the routing protocols used on the network worked. The problem was he assumed that any employee only knew a subset of what he knew, so that employee's only value-add was saving him time, not any expertise. Something like: "I know your job better than you do, now get out of my way so I fix the problem quickly and get back to important stuff that you don't understand"
      - ryandrake 11y ago
        > The problem was he assumed that any employee only knew a subset of what he knew, so that employee's only value-add was saving him time, not any expertise. Wow, if I didn't know any better, I'd bet we worked at the same company! However, my guess is that the above attitude is commonly found among type-A founders. They would not have gotten to where they are without a healthy dose of arrogance.
- biot 11y ago
  > So why do you feel the need to over-ride your product managers on > product direction, feature set, user acquisition plans, retention > strategy, etc? As long as the CEO isn't micro-managing, why wouldn't they intervene on these topics if the product managers are going in a direction the CEO doesn't want to take the company just yet? The CEO's job is to set the strategy and vision; the manager's job is to execute. If you've hired competent managers, how they execute is up to them. What and why they execute is the responsibility of the CEO and, if the managers are off-course, the CEO must sometimes override them to line things back up with the vision and strategy.
  - ryandrake 11y ago
    Overall strategy: fine. Vision: fine. Feature-by-feature product definition? Not so much. If you, as CEO, want to do that, great--just don't go out and hire product management professionals. They'll have nothing to do, with you doing their job for them. They'll eventually leave out of boredom and/or frustration.
    - biot 11y ago
      I suspect we're mostly in agreement. Going through and defining every single feature is micromanagement. However, product direction, feature set, and covering user acquisition plans and retention strategy are places where the CEO (a primary stakeholder) must have input and override where necessary; I see that as non-negotiable. There's certainly a balance that must be struck somewhere between the CEO micromanaging all the decisions and the product managers operating in a vacuum.
mpdehaan2 11y ago
"A friend of mine led his company from nothing to over $1 billion in revenue / This worked brilliantly up to about 500 employees." I'm often confused by the apparent VC assumption with excessive growth and not building right-sized companies sometimes. Is this failure? To me, this is a great time to say "I'm making lots of money, my employees are making lots of money, and it's ok to be at this size and there's nothing wrong" It might have been a great time to stop at 100, or even 25, and you could have had a fantastic time, doing one thing supremely well, too. There's honor in that. I think the pressure to grow to unicorn sizes inherrent in VC things could often result in weird business choices. It's important to remember the VC is motivated by the final selling price, and big companies sell more -- but that's totally not failure.
- AceJohnny2 11y ago
  Yes, you've come across the core criticism of the VC business model. They fund 500 startups in the hopes that one of them goes big. Not big, gigantic! Humonguous! The next Google! Facebook! Dropbox! They need such a huge return for their business model to work (do you know how expensive it is to fund 500 pie-in-the-sky startups?) With that in mind, if as a founder you're happy to settle for mid-size, you're going to get into headlong conflict with VCs who've funded you (and thus own majority shares of the company), because you're failing them according to their model. You may be interested in jwz's and idlewords's comments, who frequently decry this model.
  - aaronbrethorst 11y ago
    With that in mind, if as a founder you're happy to settle for mid-size, you're going to get fired by your board. FTFY.
    - AceJohnny2 11y ago
      I wanted to avoid sounding too alarmist, but yeah :)
    - InclinedPlane 11y ago
marktangotango 11y ago
Just left an example of this, founder built a product with a core cadre of about 5 developers, after a slow first 5 years, the market caught up, and then sales doubled every year for next 5 years ($100M plus in sales). CEO was highly involved with the product, even led the dev team for several years. Unfortunately he was little more than an enthusiastic amateur when it came to technical matters, and had strong opinions on how to implement things. His team was not experienced, so didn't recognize the hole they'd dug themselves. This led to a lava layered mess that failed to scale to the largest customers. Enter investors, a rewrite, and plummeting sales. CEO was replaced with a sales guy. See ya, good luck with that!
- shostack 11y ago
  Out of curiosity, how did the investors and rewrite result in plummeting sales? Was it the distraction from improving the product that let competitors get ahead? As you wrote it, it seemed the core issue was the technical debt that blocked to large customer sales/hurt retention. So if that was fixed in the rewrite (a big assumption admittedly), I'm left wondering what exactly went wrong.
  - marktangotango 11y ago
    The business was overly optimistic and misunderstood the scope the project and announced it way to soon ie at the beginning of the year. Miscommunication and lack of effective project management caused the rewrite to go way over the already unrealistic deadlines. Sales dried up as everyone was waiting for the new hot. Investors exacerbated the problem by requiring a 10% head count reduction for no other reason than "it will make everyone else work harder". With blood already in the water, and the train wreck of the rewrite, they started losing their top talent. I bailed because it was clear they missed core functionality, and supporting the new release was going to be a nightmare. And yes I did try to raise awareness of the issues I saw, to the point of shouting in meetings and being a jackass, to no avail. Note I only did that once, then checked out mentally, it's just not worth it. Should've left then, but stuck around for another year. When the release finally happened, it was situation where vp of IT had to make a deadline or lose his job, so the new release was crippled from the start. Sales went from doubling every year to falling well below the previous year. Edit: added a lot of detail.
femto 11y ago
The "The Product CEO Paradox" is applicable to anyone who leads the development of a product in a growing company, not just the CEO, but the Chief Engineer or CTO as well. It addresses the question of how to continue to make a useful contribution once the pace of development gets too rapid for a single person to keep up with.
hobs 11y ago
And hidden in the middle of the post is: Be the integrator – When Larry Page took over as CEO of Google, he spent a huge amount of his time forcing every product group to get to a common user profile and sharing paradigm. Why? Because he had to. It would never have happened without the CEO making it happen. It was nobody else’s top priority. This is a good example where a strong leader can push the entire company in (what I think even Google would now admit was) the wrong direction, and because they are so powerful, the core products can suffer.
- ohitsdom 11y ago
  I don't think the direction was wrong, I just think the implementation was awful. But your point still stands, it ended up harming the product.
  - hobs 11y ago
    I think that is a fair point and worth the distinction. G+ did not did not need to be deployed in such a heavy handed fashion, and if it hadn't there is a very strong probability G+ would be highly utilized and valued today.
graycat 11y ago
It sounds like the failings of the product CEO can be fixed with just a good COO. E.g., the OP has: > Then somewhere along the line, employees start complaining that the CEO is paying too much attention to what the employees can do better without her and not enough attention to the rest of the company. Sounds like the CEO needs a good COO for "the rest of the company". Okay. It does appear that VCs commonly miss this point. Maybe someone could guess why, with three guesses, where the first two don't count. For a 101 level lesson on deception, watch, say, Star Wars III -- that lesson is in the movie because audiences are prepared to believe that it is close enough to reality, and many in the audience have good reason to so believe. Deception, etc., go way back, at least to Shakespeare. Don't fall for it. In the OP, the recommended solution for the product CEO is to write down the vision or whatever for the product. Good. But now we are on the way back to the most rigid version of waterfall development -- documents mostly in advance for each level of the work. I tend to believe that those documents, etc. are a good, maybe nearly essential, approach for the scenario in the OP, but that approach is the polar opposite of lean and agile development. For small projects, sure, lean and agile are okay -- knock out some code, see if like it, else change it, rinse and repeat. But for significant projects, I never liked lean or agile anyway! But, let's see: The product CEO is to communicate only via a carefully prepared, written document and does prepare one or more such documents. So, now the VCs can think what that they couldn't before? Three guesses, and the first two don't count. There is: > The CEO skill set is incredibly difficult to master, Hmm ... So, how did Gates, Ellison, Page, Zuck do it? Did they have years of experience managing people, from a lemonade stand to the decorations for the homecoming dance to the college PBK awards dinner to a division in the US Army, to salesman, local sales manager, district sales manager, chief marketing officer? Nope -- none of those. What about guys in the flyover states who build businesses big enough to let them have, say, a 60 foot yacht? But, still we're talking something "incredibly difficult"? IMHO the whole OP is to make excuses for what VCs still very much like to do -- have the CEO their puppet on their strings and, then, pull the strings to get what they believe they want where firing the founding CEO is to be expected. Such firing by VCs used to be expected, and now we see that with A16Z it is again. Besides, this stuff about how much A16Z likes product CEOs looks like just misdirection to keep from talking about what A16Z really cares about: A company with traction significantly large and growing rapidly in a huge market but where the founders need/want cash and are still willing to sign a standard term sheet that has the founders suddenly go from owning 100% of the company to owning 0% and a four year vesting plan to get back some of the ownership while the BoD can fire them before their stock is vested and, really, makes the VCs' investments more valuable -- it's the VC's fiduciary responsibility to their limited partners. What is a founder to believe about a VC? What's in a blog post or what's in a term sheet and the other documents, maybe two inches thick, worked out very carefully by bright, well paid lawyers? Instead, (A) plan the project so that get to positive free cash flow early on, (B) grow just organically, that is, from retained earnings, and (C) plan the project so that have some good barriers to entry to keep out competition long enough for the growth. Guys, it looks like A16Z just wants your work, ownership, and company and wants to get those by having you sign an onerous term sheet while they give you some excuses about why you should give up your power as CEO and, thus, really, all or a lot of the financial benefit of your work. They are after the founder's money, guys.
vonnik 11y ago
Elephant in the room question: who is Ben talking about?
paulsutter 11y ago
It's important to delegate, but the one thing you can't delegate is strategy. The product founder needs to step back from the details but maintain direct involvement in the product aspects related to company strategy.
daniel-cussen 11y ago
Typo: "I thought founders were supposed to better?" You accidentally a word.

173 points by ggonweb 11y ago

OT: There is a bug in this blog that when I tried to share it on Facebook, it shows: "<a data-width="300" data-height="200" data-bop-link..."

scintill76 11y ago
I saw a shared post on Facebook the other day with the title "302 Found". I'm guessing it was taken from the standard generated HTML Apache etc. will send when redirecting. Not sure if it's more Facebook's or that site's fault.

ryandrake 11y ago

Great article. While it's great to have a CEO who stays engaged with the product even as the company scales up, it's awful when they continue to micro-manage it the way they did when the company was 10 people. You hired product managers (and perhaps product manager managers). So let them do their jobs and create great products! When you're running a 500+ person company, you're not going to go into the code and start optimizing the graph traversal algorithm--you hired smart engineers to do this. You're not going to fire up Illustrator and start drawing buttons--you have artists for this. You're not going to write the company blog yourself, or fix the automated test systems yourself, or design the marketing materials yourself. So why do you feel the need to over-ride your product managers on product direction, feature set, user acquisition plans, retention strategy, etc? Doing so de-motivates the professionals you have whose job is to do these awesomely.

mpdehaan2 11y ago
Product managers are potentially both the most helpful and potentially most damaging hire you can make, as you are ceding them control (if you do) over the very definition of the value your company provides. I think someone like Steve Jobs keeping a lot of that control there is definitely of value - in not delegating it. I've worked with a lot of PMs that were like "this dialog needs to be more orange" (one defunct startup from many years ago), and miss the picture, because they are just kind of faking it -- they are the guy with the opinion and the title, because somebody had to do it, but not really trying to make the best choice for the end user. There are some that could be exceptional. (Speaking of several different companies where I've experienced this). You need to have someone with a really good gut feel and instinct, and a lot of passion to believe in the thing they are doing. But if a PM is a "CEO of a product", that makes it the most important choice in your entire organization -- for that product. If you have one product, that's huge. An alternative approach is to get a company really involved, to where everybody on the team has input in the product, and that you source the ideas from everywhere, but yeah, somebody still has to drive it. Depending on who you talk to PM can mean like 37 different things, and no one person can concentrate on all 37 -- you do need people doing all of those things, but I think it's GREAT if the person at the helm of your company is all about the product. Else, you're just cranking size and numbers? The purpose of the company, to me, is about creating the things you believe in and want to create - the business is the means to the end, and the food on the table along the way. Approaching it from a pure business perspective is missing a lot of the fun and opportunity.
- omouse 11y ago
  >Product managers are potentially both the most helpful and potentially most damaging hire you can make, as you are ceding them control (if you do) over the very definition of the value your company provides. I think it ties into what paul graham wrote about hackers making things for users. The more layers you have between the hackers and the users, there's more potentially that your product won't be that great. I've worked with product managers who supposedly talk to the users directly and channel that into the work us developers have to do but they always seemed to just copy the competitor or only make small changes whose value is questionable. >An alternative approach is to get a company really involved, to where everybody on the team has input in the product, and that you source the ideas from everywhere, but yeah, somebody still has to drive it. It's probably better to create the position of product manager from within.
teacup50 11y ago
Unified vision matters. Peer commentator already noted Steve Jobs, but I wanted to focus on this. Steve didn't draw the buttons, design the hardware, or write the code -- but he sure as shit told the relevant teams exactly what he wanted -- often directly, with a very shallow management hierarchy -- and told them to try again if they didn't pull it off. At the same time, he did have SVP-level and VP-level people writing serious, core OS-level code themselves. They were better managers by virtue of actually understanding and having a coherent vision for what it was they were managing. If the management chain doesn't set product and marketing direction at a company whose purpose is to sell products, then what the hell are they doing in charge?
- teej 11y ago
  "At the same time, he did have SVP-level and VP-level people writing serious, core OS-level code themselves." I find this fascinating. How can I find out more? Which VPs? Which code?
  - teacup50 11y ago
    https://en.wikipedia.org/wiki/Bertrand_Serlet https://en.wikipedia.org/wiki/Bertrand_Serlet https://en.wikipedia.org/wiki/Avie_Tevanian https://en.wikipedia.org/wiki/Avie_Tevanian Bertrand wrote malloc, top, and quite a bit of code at NeXT. Avie wrote Mac Missles! in the 80s: https://www.youtube.com/watch?v=Ir8H0NuPZRU https://www.youtube.com/watch?v=Ir8H0NuPZRU ... and, well, Mach: http://www.cs.ubc.ca/~norm/508/2009W1/mach_usenix86.pdf ftp://ftp.cs.cmu.edu/project/mach/doc/unpublished/exception.ps ... etc etc. Steve Jobs relied on extremely technically capable management that didn't just hand off understanding to their staff.
slyall 11y ago
I had a CEO like that once. He was very good at giving them impression that he could do the job of anyone in the company better than that person and the only reason anyone else was hired was cause he couldn't be in 200 places at once.
- noskynethere 11y ago
  Can you elaborate a bit more? Could he not do other people's jobs, not do them as well? If he was correct in his assessment, was the issue that it affected morale? Right now, I feel I could do maybe 15 peoples jobs. Not as well, certainly unless I were full time, and so I cede control to them as long as they can explain their decisions. But often I feel like that statement, even if true, bodes poorly for the company. Is that because of how employees respond to that attitude?
  - slyall 11y ago
    He certainly couldn't do many of the technical guy's jobs as well as they could. Even if he'd run the whole thing by himself 5 years earlier he had not kept up with the new systems when everything had grown by 10x. Even for relatively simply things like helpdesk jobs he could just "make it so" since he was the CEO (eg just give customer a credit) rather than having to go though the normal procedures but his technical knowledge was still out-of-date there. He wasn't the nicest person in general. The problem was that he assumed he could jump into any situation and was as expert as any other staff member. So he'd start telling the network guys what to do to solve something (and trying to login to the routers) despite having virtually no idea how the routing protocols used on the network worked. The problem was he assumed that any employee only knew a subset of what he knew, so that employee's only value-add was saving him time, not any expertise. Something like: "I know your job better than you do, now get out of my way so I fix the problem quickly and get back to important stuff that you don't understand"
    - ryandrake 11y ago
      > The problem was he assumed that any employee only knew a subset of what he knew, so that employee's only value-add was saving him time, not any expertise. Wow, if I didn't know any better, I'd bet we worked at the same company! However, my guess is that the above attitude is commonly found among type-A founders. They would not have gotten to where they are without a healthy dose of arrogance.
biot 11y ago
> So why do you feel the need to over-ride your product managers on > product direction, feature set, user acquisition plans, retention > strategy, etc? As long as the CEO isn't micro-managing, why wouldn't they intervene on these topics if the product managers are going in a direction the CEO doesn't want to take the company just yet? The CEO's job is to set the strategy and vision; the manager's job is to execute. If you've hired competent managers, how they execute is up to them. What and why they execute is the responsibility of the CEO and, if the managers are off-course, the CEO must sometimes override them to line things back up with the vision and strategy.
- ryandrake 11y ago
  Overall strategy: fine. Vision: fine. Feature-by-feature product definition? Not so much. If you, as CEO, want to do that, great--just don't go out and hire product management professionals. They'll have nothing to do, with you doing their job for them. They'll eventually leave out of boredom and/or frustration.
  - biot 11y ago
    I suspect we're mostly in agreement. Going through and defining every single feature is micromanagement. However, product direction, feature set, and covering user acquisition plans and retention strategy are places where the CEO (a primary stakeholder) must have input and override where necessary; I see that as non-negotiable. There's certainly a balance that must be struck somewhere between the CEO micromanaging all the decisions and the product managers operating in a vacuum.

mpdehaan2 11y ago

"A friend of mine led his company from nothing to over $1 billion in revenue / This worked brilliantly up to about 500 employees." I'm often confused by the apparent VC assumption with excessive growth and not building right-sized companies sometimes. Is this failure? To me, this is a great time to say "I'm making lots of money, my employees are making lots of money, and it's ok to be at this size and there's nothing wrong" It might have been a great time to stop at 100, or even 25, and you could have had a fantastic time, doing one thing supremely well, too. There's honor in that. I think the pressure to grow to unicorn sizes inherrent in VC things could often result in weird business choices. It's important to remember the VC is motivated by the final selling price, and big companies sell more -- but that's totally not failure.

AceJohnny2 11y ago
Yes, you've come across the core criticism of the VC business model. They fund 500 startups in the hopes that one of them goes big. Not big, gigantic! Humonguous! The next Google! Facebook! Dropbox! They need such a huge return for their business model to work (do you know how expensive it is to fund 500 pie-in-the-sky startups?) With that in mind, if as a founder you're happy to settle for mid-size, you're going to get into headlong conflict with VCs who've funded you (and thus own majority shares of the company), because you're failing them according to their model. You may be interested in jwz's and idlewords's comments, who frequently decry this model.
- aaronbrethorst 11y ago
  With that in mind, if as a founder you're happy to settle for mid-size, you're going to get fired by your board. FTFY.
  - AceJohnny2 11y ago
    I wanted to avoid sounding too alarmist, but yeah :)
  - InclinedPlane 11y ago

marktangotango 11y ago

Just left an example of this, founder built a product with a core cadre of about 5 developers, after a slow first 5 years, the market caught up, and then sales doubled every year for next 5 years ($100M plus in sales). CEO was highly involved with the product, even led the dev team for several years. Unfortunately he was little more than an enthusiastic amateur when it came to technical matters, and had strong opinions on how to implement things. His team was not experienced, so didn't recognize the hole they'd dug themselves. This led to a lava layered mess that failed to scale to the largest customers. Enter investors, a rewrite, and plummeting sales. CEO was replaced with a sales guy. See ya, good luck with that!

shostack 11y ago
Out of curiosity, how did the investors and rewrite result in plummeting sales? Was it the distraction from improving the product that let competitors get ahead? As you wrote it, it seemed the core issue was the technical debt that blocked to large customer sales/hurt retention. So if that was fixed in the rewrite (a big assumption admittedly), I'm left wondering what exactly went wrong.
- marktangotango 11y ago
  The business was overly optimistic and misunderstood the scope the project and announced it way to soon ie at the beginning of the year. Miscommunication and lack of effective project management caused the rewrite to go way over the already unrealistic deadlines. Sales dried up as everyone was waiting for the new hot. Investors exacerbated the problem by requiring a 10% head count reduction for no other reason than "it will make everyone else work harder". With blood already in the water, and the train wreck of the rewrite, they started losing their top talent. I bailed because it was clear they missed core functionality, and supporting the new release was going to be a nightmare. And yes I did try to raise awareness of the issues I saw, to the point of shouting in meetings and being a jackass, to no avail. Note I only did that once, then checked out mentally, it's just not worth it. Should've left then, but stuck around for another year. When the release finally happened, it was situation where vp of IT had to make a deadline or lose his job, so the new release was crippled from the start. Sales went from doubling every year to falling well below the previous year. Edit: added a lot of detail.

femto 11y ago

The "The Product CEO Paradox" is applicable to anyone who leads the development of a product in a growing company, not just the CEO, but the Chief Engineer or CTO as well. It addresses the question of how to continue to make a useful contribution once the pace of development gets too rapid for a single person to keep up with.

hobs 11y ago

And hidden in the middle of the post is: Be the integrator – When Larry Page took over as CEO of Google, he spent a huge amount of his time forcing every product group to get to a common user profile and sharing paradigm. Why? Because he had to. It would never have happened without the CEO making it happen. It was nobody else’s top priority. This is a good example where a strong leader can push the entire company in (what I think even Google would now admit was) the wrong direction, and because they are so powerful, the core products can suffer.

ohitsdom 11y ago
I don't think the direction was wrong, I just think the implementation was awful. But your point still stands, it ended up harming the product.
- hobs 11y ago
  I think that is a fair point and worth the distinction. G+ did not did not need to be deployed in such a heavy handed fashion, and if it hadn't there is a very strong probability G+ would be highly utilized and valued today.

graycat 11y ago

It sounds like the failings of the product CEO can be fixed with just a good COO. E.g., the OP has: > Then somewhere along the line, employees start complaining that the CEO is paying too much attention to what the employees can do better without her and not enough attention to the rest of the company. Sounds like the CEO needs a good COO for "the rest of the company". Okay. It does appear that VCs commonly miss this point. Maybe someone could guess why, with three guesses, where the first two don't count. For a 101 level lesson on deception, watch, say, Star Wars III -- that lesson is in the movie because audiences are prepared to believe that it is close enough to reality, and many in the audience have good reason to so believe. Deception, etc., go way back, at least to Shakespeare. Don't fall for it. In the OP, the recommended solution for the product CEO is to write down the vision or whatever for the product. Good. But now we are on the way back to the most rigid version of waterfall development -- documents mostly in advance for each level of the work. I tend to believe that those documents, etc. are a good, maybe nearly essential, approach for the scenario in the OP, but that approach is the polar opposite of lean and agile development. For small projects, sure, lean and agile are okay -- knock out some code, see if like it, else change it, rinse and repeat. But for significant projects, I never liked lean or agile anyway! But, let's see: The product CEO is to communicate only via a carefully prepared, written document and does prepare one or more such documents. So, now the VCs can think what that they couldn't before? Three guesses, and the first two don't count. There is: > The CEO skill set is incredibly difficult to master, Hmm ... So, how did Gates, Ellison, Page, Zuck do it? Did they have years of experience managing people, from a lemonade stand to the decorations for the homecoming dance to the college PBK awards dinner to a division in the US Army, to salesman, local sales manager, district sales manager, chief marketing officer? Nope -- none of those. What about guys in the flyover states who build businesses big enough to let them have, say, a 60 foot yacht? But, still we're talking something "incredibly difficult"? IMHO the whole OP is to make excuses for what VCs still very much like to do -- have the CEO their puppet on their strings and, then, pull the strings to get what they believe they want where firing the founding CEO is to be expected. Such firing by VCs used to be expected, and now we see that with A16Z it is again. Besides, this stuff about how much A16Z likes product CEOs looks like just misdirection to keep from talking about what A16Z really cares about: A company with traction significantly large and growing rapidly in a huge market but where the founders need/want cash and are still willing to sign a standard term sheet that has the founders suddenly go from owning 100% of the company to owning 0% and a four year vesting plan to get back some of the ownership while the BoD can fire them before their stock is vested and, really, makes the VCs' investments more valuable -- it's the VC's fiduciary responsibility to their limited partners. What is a founder to believe about a VC? What's in a blog post or what's in a term sheet and the other documents, maybe two inches thick, worked out very carefully by bright, well paid lawyers? Instead, (A) plan the project so that get to positive free cash flow early on, (B) grow just organically, that is, from retained earnings, and (C) plan the project so that have some good barriers to entry to keep out competition long enough for the growth. Guys, it looks like A16Z just wants your work, ownership, and company and wants to get those by having you sign an onerous term sheet while they give you some excuses about why you should give up your power as CEO and, thus, really, all or a lot of the financial benefit of your work. They are after the founder's money, guys.

vonnik 11y ago

Elephant in the room question: who is Ben talking about?

paulsutter 11y ago

It's important to delegate, but the one thing you can't delegate is strategy. The product founder needs to step back from the details but maintain direct involvement in the product aspects related to company strategy.

daniel-cussen 11y ago

Typo: "I thought founders were supposed to better?" You accidentally a word.