Another plausible reason is being in a non-techy market, because you run headlong into the 1/10/89 activity pyramid except it is closer to 1/10/989: people who can't tell the difference between the green Googles and the blue Googles don't often start blogs to talk about their abiding love for invoicing software, and they don't come to websites directly because they have learned a way which works to access your Googles and that other way with the typing a website email address never really worked that well. (They still can make great customers, though.)
We wrote the post because we see a lot of people focusing on traffic when it makes more sense to tighten and work on the funnel. We have been guilty of this too :)
And yes, this doesn't really apply to non-technical markets, remember the "Facebook login" debacle? There's a whole slew of users that don't use the address bar, they just use the Googles.
This is an important thing to keep in mind, since as an SEO you'll typically be wanting to influence folks in those first two categories with the eventual intention of increasing your reach among folks in the last category who only want to pull out their credit cards.
If you want to get into it, you can do a lot of sociological work on how creators are quirky fish and how/why to optimize for their preferences. (For example, there is a strong anticommercialist streak among many influential folks on the Internet, so a core skill for businesses is not looking like they're just in this to separate you from your money while still successfully separating people from their money.)
It's like when you see those twitter statistics that suggest only a fraction of the userbase actually Tweets, and the rest are consuming.
This is relevant with regards to the 'Why aren't people talking about you?' angle that they bring up.
This group is only likely to become larger once ICANN start allowing registration of new gTLDs (http://en.wikipedia.org/wiki/Generic_top-level_domain#New_to...)
My sites typically get 75% of their traffic from search engines. I know that it could be all gone tomorrow, so I've worked on developing new sources of referrals. I still haven't found anything that delivers traffic as consistently as SEO.
In fact, generating advertising revenue from long tail traffic is a big part of my business. I don't feel like I'm in a death trap. Implementing the advertising is actually one of the best things that has happened to my business in a long time.
My company is a content business, not a software business, but I've never worried that I'm over-optimizing. If anything, I feel like I spend too little time micro-optimizing, and I spend a lot of time focusing on SEO.
Aaron Wall calls it Self Reinforcing Authority: http://www.seobook.com/archives/002033.shtml
A good user experience can also offer compounding returns, but it's hard to test and improve user experience without users. SEO can provide a steady, free stream of users that enables you to test changes large and small at a statistically significant level. That's really valuable.
for a smaller startup that is still trying to develop their product and push out an MVP, their best bang for the buck is going to be hitting the effective basics until they're ready and able to spend more time on more in-depth things.
In many ways having a good understanding of SEO is fantastic. When you understand what search engines want, and you know how to give it to them, you can get traffic that your competitors have no idea even exists. The small guy that understands SEO has a tremendous advantage over the big guy that doesn't.
However, for a small software company, I think it can be important to avoid going overboard. If you're making software (or a web service or whatever), SEO is just one piece of the puzzle, and there are only so many hours in the day. Generally only 24. Keeping close track of rankings, keywords, inbound links, and new linking opportunities is time consuming, and, when you're operating in just one niche, you can reach a point of diminishing returns. For niche products, there's only so much relevant traffic out there.
If you have multiple SEO-savvy competitors competing for your keywords, then maybe you do need to focus more time on day-to-day SEO, to keep the traffic rolling in. But many niche software businesses are doing something new, and actually don't have all that much competition for the keywords that line up well with what they're offering.
I think that often the best way for small software businesses to approach SEO it is to bake it into the systems they create, then leave those systems to do their job, building links, awareness and search traffic that will increase over time without direct involvement. By "systems" I don't mean scrapers and bots, I mean systems that encourage people to market your site for you (including building you good, natural links), and maybe systems that generate good link-worthy content for you as well.
But the problem is that Google controls traffic to that extent. Almost any website if asked honestly will say that at least 60% and more likely 80% or more of their traffic comes from Google.
