8 ms·

Small Projects, Big Companies

217 points by marceee0901 7y ago

mwilcox 7y ago
I don't think you could come up with a better critique of the problems with silicon valley today than "this product wouldn’t have billions of users, but you could charge the elite few quite a bit for using it."
- arman_ashrafian 7y ago
  could you elaborate
  - bransonf 7y ago
    Not OP, but I think the point here is that very few end users have a say in the tools they use. You sell the big company on your software, one or a few elites like it, and they buy thousands of licenses for all of their employees.
  - buboard 7y ago
    it doesnt matter what you sell, but who you sell it to
  - thundergolfer 7y ago
    Things are getting built for the 1% because they have the economic power, but I'd imagine OP thinks this is undemocratic (which it is) and results in many people becoming underserved by software companies (very likely).
    - anchpop 7y ago
      What is getting built for specifically the 1% (or .1%, etc.)? The only service I can think of is Magic. Additionally, what markets are underserved? This is an area I'm interested in but don't know much about (except from my own anecdotal experience like the fact that comparing costs of healthcare from different practitioners being prohibitively difficult)
      - thundergolfer 7y ago
        > What is getting built for specifically the 1% Should clarify that by "specifically for" I'd mean that the 1% have all the money and thus practically all the decision making power. An example of the distinction would be that though everybody engages with AdTech software the software is built for the 1% to make them money. The 1% isn't a demographic that's targeted by some SaaS in this case, it's a way of naming the rich who create the demand for a lot of software (they use the software to make money) and control the (high-level) implementation of most software in the world. What's Magic?
        anchpop 7y ago
        It's a service where you can hire an on-demand personal assistant for $30/hour, https://getmagic.com/ https://getmagic.com/. I never saw the appeal.
goatherders 7y ago
I love that the post points out the importance of selling to executives instead of users. As much as users hate most of their tools, they have very little say on what gets bought. Executives make those decisions and no amount of grousing or changing from internal users will change that. Solutions that sell upwards in an org (yammer for example) are very much the exception.
- ftio 7y ago
  This is spot on. I'm working on a blog post about a superset of this: the product IS the go-to-market. Messaging, sales/marketing strategy, pricing are just as much The Product as any feature. Ideally you maximize value/experience for users while you figure out how to get execs to buy your thing, but the sad state of most B2B software is a testament to how insanely hard it is to be good at both.
  - saadalem 7y ago
    Can you post a link to your blog ?
    - musingsole 7y ago
      https://www.ft.io/blog/ https://www.ft.io/blog/
- softwaredoug 7y ago
  Yes though it’s more complicated than this. Executives often don’t have time for every purchasing decision. And a lot of “CEO plays golf with other CEO” don’t lead to purchasing decisions that gain traction with anyone at the client that has to work in the problem domain, and can lead to dissapointing results for the purchasing org. What you want is the right level of decision making for what you’re selling. Sometimes it’s a skip level manager with budgetary authority and expertise to champion purchasing. Other times it’s many different stakeholders up and down the org chart (from end users to execs). In my experience you won’t get exec buy in if you don’t have their subordinates and others championing you as the best choice. If the in house expert in the domain isn’t interested, you won’t succeed with the executive.
chime 7y ago
> The golden dataset everyone wants is the org chart. Who reports to whom. Who is new. Who has been at the company for a while. Find a way to get this data and sell it to companies. I suspect you’ll have many takers. Please don’t. Please. Really. Please don’t. Training my users to fend off social engineering attacks is hard enough as is. Please don’t use your talents in this way.
- wheelerwj 7y ago
  This is basic B2B sales....
- cubecul 7y ago
  The Information is doing this: https://www.theinformation.com/org-charts https://www.theinformation.com/org-charts
  - thundergolfer 7y ago
    Interesting, that looks like exactly what they're talking about. Are you a subscriber, and if so how are the charts?
    - tony8000 7y ago
      It's mostly the who reports to whom part and goes about 1 to 4 levels below the CEO.
- thestepafter 7y ago
  I think this is the exact point of the article. Executives don’t think like you, so if you want to make it big with your software, you have to change how you think. Even if it means annoying middle management and their direct reports.
  - trimbo 7y ago
    That might be the point of the article, but the point of the grandparent is that the only method to get this data is to be seedy as hell. And that we should aspire to use our talents for better things.
tarr11 7y ago
The thesis seems to “build enterprise software” because those people have a budget. I enjoy reading people’s startup ideas. But this is a pretty uninspired list - (“Better Google Groups”, “Better Zendesk”, etc) There are thousands of solutions to these problems out there. Not sure what makes them more ripe for a startup then bill splitting or pizza.
- tylerrobinson 7y ago
  I think you said exactly why these are more ripe. Your customer has a budget for it.
- MetalGuru 7y ago
  It’s also incredibly hard to sell to enterprises as a startup. Building a better mousetrap is not all it takes, which explains why these solutions are still widely used despite their deficiencies. People are building better solutions all the time that go nowhere.
saadalem 7y ago
Can someone elaborate on "Backchannel.app" thing ?
- scarejunba 7y ago
  It's the common use-case: you don't know whether to trust someone. If you ever know anyone who is in danger of being hired by a friend, your friend will ask you how it was working with that person. This is usually a high-signal source because you're not going to lie to your friend, your recommendation or anti-recommendations are things you want to be meaningful, and because you have direct evidence. They want to scale this. I've discussed this same thing with friends before but we couldn't find a way to keep it clean. This is also a powerful advantage of a good network. You hear someone's leaving some job through the grapevine and if it's accompanied with positive feedback you hire this person. Or you're looking for a person and you ask your network for the guy who can do it. That's how I got my current job.
  - phkahler 7y ago
    But once you automate the backchannnel, it becomes impersonal. If someone is vindictive they can ruin careers through such a system. If you dont actually know a person, how can you trust their recommendation?
    - scarejunba 7y ago
      Well, we couldn’t figure out how to make things work and decided not to pursue. Maybe the next guy will have better luck.
- 5cott0 7y ago
  Something that is at best bordering on unethical at worst a material breach of candidate confidentiality.
macinjosh 7y ago
For #7 there is the great webmerge https://www.webmerge.me https://www.webmerge.me
arciini 7y ago
> Legal-Doc-to-Google-Form > Every founder goes through an identical process: downloading Microsoft Word, changing variables in a legal template, opening it in Preview, attaching their signature, and sending it over to the counterparty. Automate this process. Offer templates for common documents (company formation, SAFEs, etc). In time, you might become the store-of-record for company documents. My friend was working on a product that does this called Formswift! I can upload a PDF, have it convert to a set of form fields, and email it out pretty quickly. It's not the most glamorous-looking startup, but it works really well for the mom-and-pop small businesses that are their usual clientele. Take a look: https://formswift.com/self-serve-builder https://formswift.com/self-serve-builder
CoderCV 7y ago
We are building better zendesk here - https://provlem.com https://provlem.com Still, many to go, but we have started to build better zendesk already.
coffeefirst 7y ago
Number 4 is ethical quicksand. Please don’t build a machine that guesses personality traits based on personal data. It won’t end well.
- curtisgiddings 7y ago
  I'm considering submitting a project that would fit well under this category. Have any ideas around what could be done to mitigate the involved risks?
  - unlinked_dll 7y ago
    stop?
    - adnzzzzZ 7y ago
      If you believe yourself to be an ethical person then you want to build companies like those yourself rather than stop, since you can be sure that you'll do your best to handle data properly while providing the world with that service. In the case of the Big 5 there are not many companies I know of that are providing the service as proposed in the article so you would have a chance to set standards and expectations around what is or isn't acceptable.
      - bathtub365 7y ago
        Doesn’t this assume that it’s ethical to use data-driven psychometric profiling and targeting with the aim to increase profit?
        curtisgiddings 7y ago
        What if those originally submitting the data could share in the profits?
        knolax 7y ago
robbiemitchell 7y ago
Re: Zendesk It's not just that ticketing products won't build next-gen reporting with thematic monitoring and analysis, it's that big companies have more than one source of customer conversations. Actually answering the questions that matter when it comes to customer contact -- "what's driving specific outcomes, and what about them has changed over time" -- needs to be done across all sources. The company I work at (frame.ai) is an add-on with built-in connectors for all the help desks like Zendesk, Intercom, Service Cloud, Help Scout, etc. for this reason.
lifeisstillgood 7y ago
"Make something managers want. " The PG quote reformulated...
netcan 7y ago
The point on selling to managers is the big one, imo. Not users not the company, at least not directly. While the "decision maker" is not the user, they may be the project lead/implementor. Certain managers are looking for low risk, high visibility projects & notable successes. They're expending a sort of social capital within their company, putting themselves on the line. Those are the managers that need to like your thing. Companies selling to these people need to focus on those individuals, and figure out what the hell "costs," are, irl, to them. It's not money
dnautics 7y ago
It's kind of disappointing that the pioneer "RFA" projects are all "more of the same" SV. I don't see anything audacious , like "build a chip that competes with Nvidia" or "build a pharmaceutical model that improves over what we have", or "find a way to do high quality journalism without a paywall". For that matter, I don't see any inkling that any of these ideas have anything in particular to do with their self-styled goal of helping outsiders get in. For reference, here is what their main page advertised as what they seek: "We fund projects and startups built by ambitious outsiders... Apply with any type of project you need help with. It could be a company, physics research, journalism, or art..." The RFA is basically a message, "but really, this is what we want" I guess it's easier to talk big than take risks with big money.
- zomglings 7y ago
  It may also be a case of different strategies for different channels. If I were a VC, I might prefer to solicit these kinds of solid, boring business ideas in an RFA and source moonshot ideas through my personal network. If I looked for moonshots from my open application process, I would probably end up wading through a lot of applications that would be time-consuming to vet.
  - dnautics 7y ago
    That's what I would do as a VC, too, but then I wouldn't premise the advertisement of the open application for my accelerator on something misleading.
joelrunyon 7y ago
Number 7 seems like a no-brainer. Is there a reason this hasn't been done yet?
- thestepafter 7y ago
  Pretty sure LegalZoom does this?
kilotaras 7y ago
Number 7 Legal-Doc-to-Google-Form is being done by Axdraft (YC W19) Disclaimer: I'm friends with one of the cofounders.
buzzkillington 7y ago
> 3. Backchannel.app Having a hard time seeing this turn into anything but linkedin wankery. The people who could give you honest responses either don't care enough to or will start a 'you scratch my back, I'll scratch yours'. > 5. Org Chart As a Service Prepared to get sued out of existence. I was at a finance firm where all external emails were sent to both legal and security. I had to redo security training because I didn't report a phone call I got from an external source that shouldn't have had my details. And god help you if you do any of those ideas to EU citizens.
cj 7y ago
> 7. Legal-Doc-to-Google-Form Shout out to Clerky who has been doing this wonderfully since 2011.
alex-wallish 7y ago
I'm curious who #6 would be valuable for. It seems like a cool thing to make, but who is the customer?

217 points by marceee0901 7y ago

I don't think you could come up with a better critique of the problems with silicon valley today than "this product wouldn’t have billions of users, but you could charge the elite few quite a bit for using it."

arman_ashrafian 7y ago
could you elaborate
- bransonf 7y ago
  Not OP, but I think the point here is that very few end users have a say in the tools they use. You sell the big company on your software, one or a few elites like it, and they buy thousands of licenses for all of their employees.
- buboard 7y ago
  it doesnt matter what you sell, but who you sell it to
- thundergolfer 7y ago
  Things are getting built for the 1% because they have the economic power, but I'd imagine OP thinks this is undemocratic (which it is) and results in many people becoming underserved by software companies (very likely).
  - anchpop 7y ago
    What is getting built for specifically the 1% (or .1%, etc.)? The only service I can think of is Magic. Additionally, what markets are underserved? This is an area I'm interested in but don't know much about (except from my own anecdotal experience like the fact that comparing costs of healthcare from different practitioners being prohibitively difficult)
    - thundergolfer 7y ago
      > What is getting built for specifically the 1% Should clarify that by "specifically for" I'd mean that the 1% have all the money and thus practically all the decision making power. An example of the distinction would be that though everybody engages with AdTech software the software is built for the 1% to make them money. The 1% isn't a demographic that's targeted by some SaaS in this case, it's a way of naming the rich who create the demand for a lot of software (they use the software to make money) and control the (high-level) implementation of most software in the world. What's Magic?
      - anchpop 7y ago
        It's a service where you can hire an on-demand personal assistant for $30/hour, https://getmagic.com/ https://getmagic.com/. I never saw the appeal.

goatherders 7y ago

I love that the post points out the importance of selling to executives instead of users. As much as users hate most of their tools, they have very little say on what gets bought. Executives make those decisions and no amount of grousing or changing from internal users will change that. Solutions that sell upwards in an org (yammer for example) are very much the exception.

ftio 7y ago
This is spot on. I'm working on a blog post about a superset of this: the product IS the go-to-market. Messaging, sales/marketing strategy, pricing are just as much The Product as any feature. Ideally you maximize value/experience for users while you figure out how to get execs to buy your thing, but the sad state of most B2B software is a testament to how insanely hard it is to be good at both.
- saadalem 7y ago
  Can you post a link to your blog ?
  - musingsole 7y ago
    https://www.ft.io/blog/ https://www.ft.io/blog/
softwaredoug 7y ago
Yes though it’s more complicated than this. Executives often don’t have time for every purchasing decision. And a lot of “CEO plays golf with other CEO” don’t lead to purchasing decisions that gain traction with anyone at the client that has to work in the problem domain, and can lead to dissapointing results for the purchasing org. What you want is the right level of decision making for what you’re selling. Sometimes it’s a skip level manager with budgetary authority and expertise to champion purchasing. Other times it’s many different stakeholders up and down the org chart (from end users to execs). In my experience you won’t get exec buy in if you don’t have their subordinates and others championing you as the best choice. If the in house expert in the domain isn’t interested, you won’t succeed with the executive.

chime 7y ago

> The golden dataset everyone wants is the org chart. Who reports to whom. Who is new. Who has been at the company for a while. Find a way to get this data and sell it to companies. I suspect you’ll have many takers. Please don’t. Please. Really. Please don’t. Training my users to fend off social engineering attacks is hard enough as is. Please don’t use your talents in this way.

wheelerwj 7y ago
This is basic B2B sales....
cubecul 7y ago
The Information is doing this: https://www.theinformation.com/org-charts https://www.theinformation.com/org-charts
- thundergolfer 7y ago
  Interesting, that looks like exactly what they're talking about. Are you a subscriber, and if so how are the charts?
  - tony8000 7y ago
    It's mostly the who reports to whom part and goes about 1 to 4 levels below the CEO.
thestepafter 7y ago
I think this is the exact point of the article. Executives don’t think like you, so if you want to make it big with your software, you have to change how you think. Even if it means annoying middle management and their direct reports.
- trimbo 7y ago
  That might be the point of the article, but the point of the grandparent is that the only method to get this data is to be seedy as hell. And that we should aspire to use our talents for better things.

tarr11 7y ago

The thesis seems to “build enterprise software” because those people have a budget. I enjoy reading people’s startup ideas. But this is a pretty uninspired list - (“Better Google Groups”, “Better Zendesk”, etc) There are thousands of solutions to these problems out there. Not sure what makes them more ripe for a startup then bill splitting or pizza.

tylerrobinson 7y ago
I think you said exactly why these are more ripe. Your customer has a budget for it.
MetalGuru 7y ago
It’s also incredibly hard to sell to enterprises as a startup. Building a better mousetrap is not all it takes, which explains why these solutions are still widely used despite their deficiencies. People are building better solutions all the time that go nowhere.

saadalem 7y ago

Can someone elaborate on "Backchannel.app" thing ?

scarejunba 7y ago
It's the common use-case: you don't know whether to trust someone. If you ever know anyone who is in danger of being hired by a friend, your friend will ask you how it was working with that person. This is usually a high-signal source because you're not going to lie to your friend, your recommendation or anti-recommendations are things you want to be meaningful, and because you have direct evidence. They want to scale this. I've discussed this same thing with friends before but we couldn't find a way to keep it clean. This is also a powerful advantage of a good network. You hear someone's leaving some job through the grapevine and if it's accompanied with positive feedback you hire this person. Or you're looking for a person and you ask your network for the guy who can do it. That's how I got my current job.
- phkahler 7y ago
  But once you automate the backchannnel, it becomes impersonal. If someone is vindictive they can ruin careers through such a system. If you dont actually know a person, how can you trust their recommendation?
  - scarejunba 7y ago
    Well, we couldn’t figure out how to make things work and decided not to pursue. Maybe the next guy will have better luck.
5cott0 7y ago
Something that is at best bordering on unethical at worst a material breach of candidate confidentiality.

macinjosh 7y ago

For #7 there is the great webmerge https://www.webmerge.me https://www.webmerge.me

arciini 7y ago

> Legal-Doc-to-Google-Form > Every founder goes through an identical process: downloading Microsoft Word, changing variables in a legal template, opening it in Preview, attaching their signature, and sending it over to the counterparty. Automate this process. Offer templates for common documents (company formation, SAFEs, etc). In time, you might become the store-of-record for company documents. My friend was working on a product that does this called Formswift! I can upload a PDF, have it convert to a set of form fields, and email it out pretty quickly. It's not the most glamorous-looking startup, but it works really well for the mom-and-pop small businesses that are their usual clientele. Take a look: https://formswift.com/self-serve-builder https://formswift.com/self-serve-builder

CoderCV 7y ago

We are building better zendesk here - https://provlem.com https://provlem.com Still, many to go, but we have started to build better zendesk already.

coffeefirst 7y ago

Number 4 is ethical quicksand. Please don’t build a machine that guesses personality traits based on personal data. It won’t end well.

curtisgiddings 7y ago
I'm considering submitting a project that would fit well under this category. Have any ideas around what could be done to mitigate the involved risks?
- unlinked_dll 7y ago
  stop?
  - adnzzzzZ 7y ago
    If you believe yourself to be an ethical person then you want to build companies like those yourself rather than stop, since you can be sure that you'll do your best to handle data properly while providing the world with that service. In the case of the Big 5 there are not many companies I know of that are providing the service as proposed in the article so you would have a chance to set standards and expectations around what is or isn't acceptable.
    - bathtub365 7y ago
      Doesn’t this assume that it’s ethical to use data-driven psychometric profiling and targeting with the aim to increase profit?
      - curtisgiddings 7y ago
        What if those originally submitting the data could share in the profits?
        knolax 7y ago

robbiemitchell 7y ago

Re: Zendesk It's not just that ticketing products won't build next-gen reporting with thematic monitoring and analysis, it's that big companies have more than one source of customer conversations. Actually answering the questions that matter when it comes to customer contact -- "what's driving specific outcomes, and what about them has changed over time" -- needs to be done across all sources. The company I work at (frame.ai) is an add-on with built-in connectors for all the help desks like Zendesk, Intercom, Service Cloud, Help Scout, etc. for this reason.

lifeisstillgood 7y ago

"Make something managers want. " The PG quote reformulated...

netcan 7y ago

The point on selling to managers is the big one, imo. Not users not the company, at least not directly. While the "decision maker" is not the user, they may be the project lead/implementor. Certain managers are looking for low risk, high visibility projects & notable successes. They're expending a sort of social capital within their company, putting themselves on the line. Those are the managers that need to like your thing. Companies selling to these people need to focus on those individuals, and figure out what the hell "costs," are, irl, to them. It's not money

dnautics 7y ago

It's kind of disappointing that the pioneer "RFA" projects are all "more of the same" SV. I don't see anything audacious , like "build a chip that competes with Nvidia" or "build a pharmaceutical model that improves over what we have", or "find a way to do high quality journalism without a paywall". For that matter, I don't see any inkling that any of these ideas have anything in particular to do with their self-styled goal of helping outsiders get in. For reference, here is what their main page advertised as what they seek: "We fund projects and startups built by ambitious outsiders... Apply with any type of project you need help with. It could be a company, physics research, journalism, or art..." The RFA is basically a message, "but really, this is what we want" I guess it's easier to talk big than take risks with big money.

zomglings 7y ago
It may also be a case of different strategies for different channels. If I were a VC, I might prefer to solicit these kinds of solid, boring business ideas in an RFA and source moonshot ideas through my personal network. If I looked for moonshots from my open application process, I would probably end up wading through a lot of applications that would be time-consuming to vet.
- dnautics 7y ago
  That's what I would do as a VC, too, but then I wouldn't premise the advertisement of the open application for my accelerator on something misleading.

joelrunyon 7y ago

Number 7 seems like a no-brainer. Is there a reason this hasn't been done yet?

thestepafter 7y ago
Pretty sure LegalZoom does this?

kilotaras 7y ago

Number 7 Legal-Doc-to-Google-Form is being done by Axdraft (YC W19) Disclaimer: I'm friends with one of the cofounders.

buzzkillington 7y ago

> 3. Backchannel.app Having a hard time seeing this turn into anything but linkedin wankery. The people who could give you honest responses either don't care enough to or will start a 'you scratch my back, I'll scratch yours'. > 5. Org Chart As a Service Prepared to get sued out of existence. I was at a finance firm where all external emails were sent to both legal and security. I had to redo security training because I didn't report a phone call I got from an external source that shouldn't have had my details. And god help you if you do any of those ideas to EU citizens.

cj 7y ago

> 7. Legal-Doc-to-Google-Form Shout out to Clerky who has been doing this wonderfully since 2011.

alex-wallish 7y ago

I'm curious who #6 would be valuable for. It seems like a cool thing to make, but who is the customer?