As a founder (and the website maintainer) of one of the startups "attacked", I'm not too bothered by the criticism. It is valid...we aren't showing up in all the right places on Google, and we're currently paying a buck or more per click to occasionally appear in those right places in ads. But, it's also worth noting that while we'd never turn down more traffic we also aren't experiencing significant pain due to that lack. Our site gets 1800 uniques per day, and is growing at a very respectable clip--I remember vividly celebrating our first week of 1000+ visitor days just a few months back and it won't be long before we're celebrating 2000+ visitor days. For a site that sells a product (rather than sells eyeballs), that's a nice stream of potential buyers. Actually, I sometimes chuckle when I read about sites that are selling nothing but eyeballs bragging about 100,000 monthly pageviews (or whatever) when our product site does significantly more than that (and our four sites combined see an order of magnitude more traffic than that).

We consider our biggest website problems to be conversion rather than bringing the eyeballs in...so even though I readily admit that we're weaker in the engines than we'd like to be, my current focus with the website is on getting those people who come to our site excited enough to buy our product (or, even better, ask their hosting provider to buy it for them), and better serving those customers that have bought.

So, I don't know that we're really a good example of a truckload of SEO fail, even if our SEO results are poorly, since we don't actually live or die by search engine traffic. Only recently did the majority of our incoming traffic shift from our Open Source project pages to other sources, for example, and those clicks from Webmin.com are still far more valuable to us than search-related clicks...they are about 10 times more likely to buy from us than natural search result users. Likewise for users of our Open Source Virtualmin version at places like Joyent, who are proving to be extremely likely to upgrade to the commercial version. This tells me that getting more people using and talking about our Open Source products is a better marketing tactic than almost anything else we could do. So, we balance our efforts, and SEO unfortunately gets put on a back-burner. (We do try hard to have good URLs, though, but mainly because I consider URLs a usability issue, and not because I care what a search engine thinks about them. I just wish Joomla didn't suck so bad for stuff like that.)

Im a virtualmin user, I love it, I use the GPL version as it's lighter and I need to squeeze out every cent of my server as memory is the big issue for my app.

The app I have running on it is purely search engine traffic driven by seo. I am top for the main 3 - 4 word terms, (suffering a slight drop with the latest end march google update) that said the next term were going after is the big one and will bring 4 fold the traffic we have now.

My position is this, if your not into seo and your doing a startup then you better have some nice friends, and hope to make the first page of TC etc. Else you will sit there with 50 hits a day and most of them will be from crawlers.

I have no 'nice friends' apart from my seo adviser. I'm lucky he is a good friend and trained me well. Now I have a site that has tremendous growth, targeted traffic that sticks about (and comes back) and absoulty zero effort wasted on pitching for editorial, advertising, other marketing, etc.

I could care less about word of mouth because it is very hit and miss. People only want to talk about you if you have upset them or if you have a real game changer. Competing or slogging it out in established markets, like music for example just doesn't get the masses enthused to enough to chat about (mind everyone will blog about you if TC did so that's a bonus of being T.cked.)

I often check my apps progress with compete and benchmark it against sites that are in the same market that are going for the big bucks, VC backing, big splashes of editorial on TC, advertising and generally hustling their way in, and while I am often behind there initial curves the encouraging thing is my traffic is constantly rising and Im nipping at there ankles as they dip after the sharp rises, while also building a large net to crawl the ocean of organic search traffic.

I also freelance, coding and SEO, and it's really amazing talking to potential clients that are working on startups, they are interested in me helping them with their apps, but when I say, look the best advice I can help you with is not coding but in getting your SEO sorted out - blank faces stare back. Just like all the examples given in this article, they are equally nonchalant.

The biggest laugh for me was a competitor who had made TC because it had been backed with 5 million USD or some other worthless currency and when I did the quick seo scan (10 secs work) the biggest laugh of my life (index=1 page).

Alas I don't expect that the startups I see here will change there focus and concentrate on SEO anytime soon, the general opinion that seems to be prevalent is that SEO is like web1, and social graphs and viral marketing will build the traffic.

That is a huge shame, I feel a huge empathy for the fellow entrepreneur's on this board, sharing many of the same trials and tribulations I feel like I also share some of the same genes, and this is nothing more than a little clan, and I want nothing more than to see my fellow clan members go out and change the world. For better and worse, that is model of creation and one bloody good reason to live.

Long live Virtualmin.

"Im a virtualmin user, I love it, I use the GPL version as it's lighter and I need to squeeze out every cent of my server as memory is the big issue for my app."

Awesome. Memory usage of GPL and Professional are roughly identical, but if you don't need the features of Professional, we'd encourage you to keep using GPL. Also check out our guide for running Virtualmin on low memory systems:

http://www.virtualmin.com/documentation/id,virtualmin_on_low...

Most memory usage is actually found in the underlying services, and not Virtualmin/Webmin proper (though Webmin with all of the Virtualmin modules cached is about 110MB...turning off that caching takes it down to 10MB).

"Alas I don't expect that the startups I see here will change there focus and concentrate on SEO anytime soon, the general opinion that seems to be prevalent is that SEO is like web1, and social graphs and viral marketing will build the traffic."

Actually, we thought hard about viral marketing and social graphs and realized that we don't stand a chance marketing via that path. Almost nothing about our product has viral characteristics, though word of mouth has been strong. We've been focusing on increasing the amount of valuable information on our network of websites (Webmin.com, Virtualmin.com, and Doxfer.com), such that searches for all sorts of common server administration problems will result in the user coming to one or more of our sites because the solution can be found there. I don't know how to measure the success of this, and it hasn't resulted in us being front page of Google for any of our keywords...but we do get several hundred visitors a day from natural search results to our forums and wikis. As I mentioned, our traffic is nothing to be ashamed of for a product site, and it's growing rapidly.

All of this discussion has made me think a lot more about the problem. So I'm going to start paying more attention and blog about the results. Starting from our current "zero SEO effort" state, I'll spend the next couple of months tuning for SEO and post a few short articles about the results. It's certainly worth some testing, if we could crank up traffic without spending ad dollars.

My problem with memory was primarily caused by poor indexing on mysql tables, that once I resolved and cleaned up my queries resolved my memory problems completely. Ref Virtualmin pro, it was the claimav that used more memory not PRO per say. However, I don't need the server for email at all, so the GPL version for me is actually a god send. As I said I really do love it and would happily pay for it. I think you should offer it as shareware for 50 usd one time license, its really good (with the install script) and is a great solution for system admins who don't want (or know how to) set up LAMP without a panel.

Im in the latter catagory, having moved over to a linode, (also highly recommended) after the app was OOM the previous server that was really lightweight and over priced, (that had cpanel) and I actually prefer Virtualmin. It's more rustic and I prefer the support from your small team.

As regards SEO traffic, I'm really glade your thinking of getting into it more. Basically you need to talk to a pro if you want to make headways, I assure you it's not just a case of having the right titles and meta tags, though of course that helps.

seekely - I'm the author of the post and am very sorry for being condescending. That wasn't my intent and on re-reading, I can see that it came across that way. I offer my apology.

I was writing a bit more emotionally than I usually do, and, as I said in the post, am much more frustrated with those who provide startup advice ignoring the power of SEO than I am with startups who don't engage in it. Startups have a million things on their plates. Those who make lists recommending marketing practices have just one task, and I think it's negligent to ignore such a valuable marketing tactic for web-based companies.

i think you're wrong about Octopart - they are optmising around individual part terms e.g. http://octopart.com/category--Integrated+Circuits--Semicondu... not for a generic "electronic parts" ranking. Their recent user data indicates this has been pretty successful - basically SEO - adding categories to search.