4 ms·

This falls into the broader category of stories "Don't build a business that is wholly dependent on another business" of which we hear variations here on Hacker

This falls into the broader category of stories "Don't build a business that is wholly dependent on another business" of which we hear variations here on Hacker News (such as "Don't build your whole business around the API of another business.") Demand Media was far too reliant on its ability to manipulate its rankings in Google. We had a similar opportunity. We built Accumulist in 2006. We wanted stories posted to Accumulist to do well at Google, but we succeeded way beyond our expectations. Something about the structure of Accumulist hit the sweet spot of whatever algorithms that Google was using during 2006. Any article posted to Accumulist jumped to the front page of Google. Accumulist had an API so outside bloggers could ping it with their stories, and generally, for any story posted to Accumulist, the page that got ranked in Google was the page on Accumulist, rather than the original page on the blog. For us, at the time, Accumulist was one of many projects we were working on. It was an unexpected hit. We did think about making it our full time focus. I was in favor of expanding it. However, my partners were against it. At one point we were working with an outside consultant whose advice I greatly respected. He said, "The business plan can not consist soley of 'Accumulist ranks high with Google.'" And that was a good argument. For businesses like Demand Media and Accumulist, I think ranking highly with a search engine can only be the beginning of a process. If that's all you've got, then quit. There has to be something else you can do, something to add to the process, something real, that gives your business independence from the other business that you are initially depending on. Otherwise I think you are building your business on sand, and its only a matter of time before it washes away.

larrys 13y ago
"The business plan can not consist soley of 'Accumulist ranks high with Google." This is really a variation of the general notion in business that you are in trouble if you only have several big customers or a few customers control most of your destiny. In this case of course google is not a customer but it's really the same thing. Eggs in a basket. Anyone with any history in business would quickly realize the problem with putting all your eggs in one basket. Certainly one of the reasons I am not a fan of SEO. Live by the sword die by the sword. Rank high at SEO and you can lose that rank. Less likely if you are paying for your advertising. Back in the day the furniture and department stores that were paying for advertising in the local paper had a good situation. A barrier to entry against competition because they paid. It wasn't free. Likewise I was very successful with the yellow pages in on past business. Any competitor could have done the same thing. But they didn't. They were to cheap to just take a chance, even with a small advertisement. I liked the fact that I paid for those ads. They worked very well. Each year we took out a bigger ad. Let's take a real world case here. Let's take grellas who gets almost certainly business from his postings on HN. Let's further assume the fact that he is to big and popular to fail and that he is not likely or even close to likely to get banned from HN. As if. But now let's add that he has managed to build his firm with more associates and rent more space as a result of being on HN (speculation on my part for the purpose of my point). Now let's say something happens and all the sudden HN is not in favor anymore. After all nothing is guaranteed. What does grellas do then (with all that overhead)? How will he quickly come up with more work? Would be much better if grellas had a formula for business that he paid for or had more control over. Not making a specific judgement or suggestion that grellas stop posting or that he is not spending his time wisely. Just drilling down to illustrate that if you get something you don't pay for you lose a certain amount of control (and a barrier to entry) that having to pay gives you.
- nl 13y ago
  That scenario isn't the same thing at all. A lawyer has recurring income, and something like HN is merely a source of new business. HN closing stops growth, but you don't lose the existing income sources. Demand Media isn't the same - "Google traffic" was the recurring business. When that went away they were screwed. (Also, I think it's inadvisable to name real people in hypothetical scenarios without their permission.)
  - larrys 13y ago
    True that a lawyer has recurring income. A "book" of business so to speak. So you are right that the situations are different. But my point is that if you are depending on something to bring you new business in a certain way you have to be careful that that thing doesn't and can't end arbitrarily. To the best of your ability at least. Or be aware of the dangers of losing it (when signing that new lease). Generally if you pay for something you have more control because the transaction is less lopsided. "Also, I think it's inadvisable to name real people in hypothetical scenarios without their permission" Why? Grellas is a public commenter here (whose identity is well known) on quite a large scale. And he is referred to by others in comments. To me it's really no different than making a comment on any commenter. Now if I said "Jane who works for PG" (just made that up) "has to be concerned if bla bla bla" that would be different perhaps. For that matter tptacek has mentioned his wife's name as "Erin" iirc. If he hadn't mentioned her it might be wrong (if I knew her name) to disclose it or to talk about her in any way or that he is even married. Or to disclose the street which he lives on. Etc. I would be curious what others think about this. Not sure I see how using grellas as an example was wrong but I'm open to others thoughts on this.
    - nl 13y ago
      "Also, I think it's inadvisable to name real people in hypothetical scenarios without their permission" Why? Grellas is a public commenter here (whose identity is well known) on quite a large scale. Because you are inserting them in a hypothetical scenario. Imagine the case where an investor Googles their name, finds your discussion of the supposed weakness in their business model and refuses to invest because of how weak their business model is. Or - perhaps even more likely - imagine software that runs sentiment analysis on people's usernames. Suddenly you have associated them with negative sentiment...
suchusername 13y ago
It seems to be an extreme case of digital sharecropping. http://www.roughtype.com/?p=1600 http://www.roughtype.com/?p=1600
rhizome 13y ago
I've never been to business school, but I have to think there is a basic tenet regarding external dependencies that Demand Media ignored, whether it was through ignorance or hubris.
- larrys 13y ago
  Would like to point out (I did go to business school) this may very well be something they teach in business school (I don't really remember) but much of business is really common sense at least with respect to entrepreneurship. (Which is why there are so many successful people that did not go to business school). Not to mention the fact that even if you do learn this in business school it might be any number of things that you learn without any degree of "how important is this anyway". Think of reading a "man" page with a million options and arguments. Someone with real life experience knows what is really important from experience to concentrate on. Anyone can memorize things.
- mathattack 13y ago
  As an MBA speaking, business school wouldn't have taught you anything beyond 3 years of reading of HN on this topic. Perhaps MBAspeak calls is not outsourcing your core competencies, but the common sense answer is not to put your entire livelihood in the hands of one external party.
aaronbrethorst 13y ago
Or you could take away the lesson of 'build something that is of real, meaningful value to users.'
- larrys 13y ago
  "real, meaningful value to users" only helps with part of the problem. The other part, depending on the whims of a big player for your traffic which is a big issue even if there is value. After all the chance of a site going down in rankings enough to impact profits is much greater than the chance that anyone is going to complain (as a group) to get that previous ranking restored. Or certainly the site operator can go in most cases and pound sand. So let's say you are "perlanswers.com" (arbitrary for example name not even registered) and you currently rank near the top. Then all the sudden you don't rank near the top and you fall to #8 or #15 which greatly hurts traffic. You can't complain and even if you have a loyal customer base "meaningful value" they really aren't going to be able to do much either. Right? Of course if you have a large group of people typing in perlanswers.com that's a different story. But how many times does that happen as opposed to the "other" way people tend to get answers?
  - wpietri 13y ago
    If you have something that's of real, meaningful value to users, I think big players are less of a problem. If I'm selling a better mousetrap and suddenly Walmart won't distribute it, there are a lot of other ways to sell mousetraps. If I have valuable content, search is one way I can get people to it, but it's far from the only way. The mistake Demand Media made was creating a lot of crappy content. They weren't delivering much user value; their real skill was search-driven eyeball arbitrage. When Google tuned their search engine to deliver more value to their users, Demand got squeezed out and had nothing to fall back on.
    - benologist 13y ago
      The mistake Demand Media made was not creating their shitty content under a veil of legitimacy - AOL calls their giant content farm "news" and their cheap writers "journalists".
cookiecaper 13y ago
Definitely agree with this attitude. There's not necessarily a problem with banking on a reasonably stable "partner" (whether they're aware of their participation or not) early on if you have a pretty good/quick/easy way to grab those coattails, but it can't be anything more than a springboard.
ghshephard 13y ago
For those who cashed out after the IPO and walked away with several 10s of millions of dollars, I would suggest that you can certainly profit (if not build a business) around "xxx ranks high with google." I'm sure that has to be an attractive outcome for a pretty large population of business people, VCs included.

larrys 13y ago

"The business plan can not consist soley of 'Accumulist ranks high with Google." This is really a variation of the general notion in business that you are in trouble if you only have several big customers or a few customers control most of your destiny. In this case of course google is not a customer but it's really the same thing. Eggs in a basket. Anyone with any history in business would quickly realize the problem with putting all your eggs in one basket. Certainly one of the reasons I am not a fan of SEO. Live by the sword die by the sword. Rank high at SEO and you can lose that rank. Less likely if you are paying for your advertising. Back in the day the furniture and department stores that were paying for advertising in the local paper had a good situation. A barrier to entry against competition because they paid. It wasn't free. Likewise I was very successful with the yellow pages in on past business. Any competitor could have done the same thing. But they didn't. They were to cheap to just take a chance, even with a small advertisement. I liked the fact that I paid for those ads. They worked very well. Each year we took out a bigger ad. Let's take a real world case here. Let's take grellas who gets almost certainly business from his postings on HN. Let's further assume the fact that he is to big and popular to fail and that he is not likely or even close to likely to get banned from HN. As if. But now let's add that he has managed to build his firm with more associates and rent more space as a result of being on HN (speculation on my part for the purpose of my point). Now let's say something happens and all the sudden HN is not in favor anymore. After all nothing is guaranteed. What does grellas do then (with all that overhead)? How will he quickly come up with more work? Would be much better if grellas had a formula for business that he paid for or had more control over. Not making a specific judgement or suggestion that grellas stop posting or that he is not spending his time wisely. Just drilling down to illustrate that if you get something you don't pay for you lose a certain amount of control (and a barrier to entry) that having to pay gives you.

nl 13y ago
That scenario isn't the same thing at all. A lawyer has recurring income, and something like HN is merely a source of new business. HN closing stops growth, but you don't lose the existing income sources. Demand Media isn't the same - "Google traffic" was the recurring business. When that went away they were screwed. (Also, I think it's inadvisable to name real people in hypothetical scenarios without their permission.)
- larrys 13y ago
  True that a lawyer has recurring income. A "book" of business so to speak. So you are right that the situations are different. But my point is that if you are depending on something to bring you new business in a certain way you have to be careful that that thing doesn't and can't end arbitrarily. To the best of your ability at least. Or be aware of the dangers of losing it (when signing that new lease). Generally if you pay for something you have more control because the transaction is less lopsided. "Also, I think it's inadvisable to name real people in hypothetical scenarios without their permission" Why? Grellas is a public commenter here (whose identity is well known) on quite a large scale. And he is referred to by others in comments. To me it's really no different than making a comment on any commenter. Now if I said "Jane who works for PG" (just made that up) "has to be concerned if bla bla bla" that would be different perhaps. For that matter tptacek has mentioned his wife's name as "Erin" iirc. If he hadn't mentioned her it might be wrong (if I knew her name) to disclose it or to talk about her in any way or that he is even married. Or to disclose the street which he lives on. Etc. I would be curious what others think about this. Not sure I see how using grellas as an example was wrong but I'm open to others thoughts on this.
  - nl 13y ago
    "Also, I think it's inadvisable to name real people in hypothetical scenarios without their permission" Why? Grellas is a public commenter here (whose identity is well known) on quite a large scale. Because you are inserting them in a hypothetical scenario. Imagine the case where an investor Googles their name, finds your discussion of the supposed weakness in their business model and refuses to invest because of how weak their business model is. Or - perhaps even more likely - imagine software that runs sentiment analysis on people's usernames. Suddenly you have associated them with negative sentiment...

suchusername 13y ago

It seems to be an extreme case of digital sharecropping. http://www.roughtype.com/?p=1600 http://www.roughtype.com/?p=1600

rhizome 13y ago

I've never been to business school, but I have to think there is a basic tenet regarding external dependencies that Demand Media ignored, whether it was through ignorance or hubris.

larrys 13y ago
Would like to point out (I did go to business school) this may very well be something they teach in business school (I don't really remember) but much of business is really common sense at least with respect to entrepreneurship. (Which is why there are so many successful people that did not go to business school). Not to mention the fact that even if you do learn this in business school it might be any number of things that you learn without any degree of "how important is this anyway". Think of reading a "man" page with a million options and arguments. Someone with real life experience knows what is really important from experience to concentrate on. Anyone can memorize things.
mathattack 13y ago
As an MBA speaking, business school wouldn't have taught you anything beyond 3 years of reading of HN on this topic. Perhaps MBAspeak calls is not outsourcing your core competencies, but the common sense answer is not to put your entire livelihood in the hands of one external party.

aaronbrethorst 13y ago

Or you could take away the lesson of 'build something that is of real, meaningful value to users.'

larrys 13y ago
"real, meaningful value to users" only helps with part of the problem. The other part, depending on the whims of a big player for your traffic which is a big issue even if there is value. After all the chance of a site going down in rankings enough to impact profits is much greater than the chance that anyone is going to complain (as a group) to get that previous ranking restored. Or certainly the site operator can go in most cases and pound sand. So let's say you are "perlanswers.com" (arbitrary for example name not even registered) and you currently rank near the top. Then all the sudden you don't rank near the top and you fall to #8 or #15 which greatly hurts traffic. You can't complain and even if you have a loyal customer base "meaningful value" they really aren't going to be able to do much either. Right? Of course if you have a large group of people typing in perlanswers.com that's a different story. But how many times does that happen as opposed to the "other" way people tend to get answers?
- wpietri 13y ago
  If you have something that's of real, meaningful value to users, I think big players are less of a problem. If I'm selling a better mousetrap and suddenly Walmart won't distribute it, there are a lot of other ways to sell mousetraps. If I have valuable content, search is one way I can get people to it, but it's far from the only way. The mistake Demand Media made was creating a lot of crappy content. They weren't delivering much user value; their real skill was search-driven eyeball arbitrage. When Google tuned their search engine to deliver more value to their users, Demand got squeezed out and had nothing to fall back on.
  - benologist 13y ago
    The mistake Demand Media made was not creating their shitty content under a veil of legitimacy - AOL calls their giant content farm "news" and their cheap writers "journalists".

cookiecaper 13y ago

Definitely agree with this attitude. There's not necessarily a problem with banking on a reasonably stable "partner" (whether they're aware of their participation or not) early on if you have a pretty good/quick/easy way to grab those coattails, but it can't be anything more than a springboard.

ghshephard 13y ago

For those who cashed out after the IPO and walked away with several 10s of millions of dollars, I would suggest that you can certainly profit (if not build a business) around "xxx ranks high with google." I'm sure that has to be an attractive outcome for a pretty large population of business people, VCs included.