Monday, November 9, 2009

SEO for Startups

I recently had a conversation with an pre-launch startup where they discussed how SEO was going to be really important for them.  They went on to describe a fairly common type of site that has some original content, but not much.  They are in a space where there is lots of search traffic, but they didn't have anything particularly interesting to say about how they were going to get ranked.

This isn't the first time I've had this same conversation.  And to me, there seems to be a lot of confusion by founders of startups around search engine optimization (SEO) and how it works for a startup.

I won't say that I'm truly an expert on SEO.  And I believe that there's a lot of randomness in results.  Still having worked with a lot of different startups and especially a lot on SEO and especially long tail seo, I'm going to say that I have a pretty good handle on what works and what doesn't.  And generally, founders overestimate the possibilities associated with SEO.

On Page SEO

Most founders talk about what I call On Page SEO.  That's structuring your pages and your site to optimize it for search.  And let's face facts, we normally are talking Google search.  If you don't know the basics here, go look at Google's Search Engine Optimization Starter Guide (PDF).  It tells you all the basics and the importance of things like:

  • Titles
  • URLs
  • Anchor Tags
  • Site Navigation
  • Content Quality

Once you've skimmed this, you'll have a pretty good idea of all the things that you should do to get your pages in good shape.

They only very briefly mention how you can effectively promote your site.  Which is also really important: off page SEO.

However, if all you do is follow the on page SEO advice, it is unlikely that you will get much traffic via search.

Off Page SEO

The other aspect of SEO is at its most basic, how many pages with decent Google Page Rank point to you using anchor text that helps you with optimization.  There's a lot more to it, but since startups generally start with zero links to them, they are way behind.  You can build Google Juice over time by getting good in bound links.  You can promote your site all over the place.

By the way – getting links in social media – mentions on twitter, links on Facebook.  They bring traffic, but have little SEO value. 

Definitely engaging with bloggers is a great way to get things going.  I'm a big fan of blogger outreach and using browse my stuff as a basis.  This can help juice your SEO. 

But it's still a hard road to climb.  Let's consider the numbers…

Search Quantity

In the same conversation about SEO for startups, the founder either tells me:

  • There's a ton of search in my space
  • I'm not sure how much search there is in my space

But most often I find that they really haven't done enough analysis.  And this is not all that hard to do.  You go use Google's Adwords Keyword Tool and you plug in your keyword and see the results.  For example, let's say I'm looking at SEO in farming.  I might get back:


So, great news.  There's a lot of search for "farms" and "farming".  But that's only a small part of the picture.  What we need to do is assess our ability to capture any of that traffic.  So I go to Google and do a search for each of these phrases.  And I do a general assessment of whether I can get into the top ten.  For farming, unless we do something really amazing with off-page SEO.  And you probably aren't.  In fact, I can't quite even imagine what you will do to rank well on these broad terms.

Oh, by the way, the numbers I'm showing you above are actually a bit misleading because I'm using "Broad" matching.  If you want the real numbers, you should go to Exact to see exact numbers put in for different terms.


So, the actual monthly numbers for farming is only 74,000 per month.  And most often specific terms drop off very fast.

Another aspect to this is that click-through rates are actually pretty low for even the top ten results.  Here's a ballpark derived from various research studies on the click through rate for positions 1-10 in Google.

1 25.0%
2 10.0%
3 7.5%
4 3.3%
5 2.7%
6 2.2%
7 1.9%
8 1.6%
9 1.6%
10 1.6%

So, if you miraculously get the number 2 position on Google for the term farming, you will get 7,400 clicks per month.  Ouch!  That's pretty low.  Of course, it will be a lot more than that because if you were able to do it for that term, then you'll get a lot of long tail searches.  Still the numbers turn pretty small fast.

And, you aren't going to rank for farming or farms.  Instead, you may rank for a much more specific set of terms.  So, what you are normally looking at is ranking for something like farm/farming issues.  Looking at the numbers for those (using Broad Match) we see:


And you can see that there might be a few thousand related searches a month.  And if you work really hard at SEO over this subset of farming, you might find yourself getting 10% of that (being very optimistic).  So, you are talking about 10-50 visitors per day.

It would have been cheaper to buy those clicks.

Value of SEO for Startups

I'm being intentionally pretty harsh about the founders who predict better results with SEO than is deserved.  I do believe that there's value in SEO for Startups.

Let's assume that you know and are willing to follow the basics:

  • Do a good job on on-page SEO.
  • Take time to build up a presence and accumulate Google Juice.
  • Blog to build credibility and create original content.  Not to mention learn a lot.
  • Look to long tail SEO opportunities, especially early.
  • Work with bloggers in your space in a smart way.

Then, the beauty of SEO is as a long-term strategy for building inexpensive traffic.  If you happen to be working in space where there's lots of search, especially lots of long tail search and even better if the SEO competition for that traffic is low, then you've got a chance to build traffic.


Eric Greenspan said...

Awesome advice. Great post Tony.

Alon said...

Great recap Tony. Though I think another good starting step is submission of media such as press releases, articles and whitepapers to free databases. It'll give you instant results for search terms that are your startup's name and provide some incoming links immediately. Not the best quality incoming links, but a good start.

Tony Karrer said...

@Eric - great to have you stop by.

@Alon - Does that give much in the way of Google Juice? I've not really found it all that helpful, but maybe I'm missing it.

Costa Rica Surfing said...

You could have included site structure and spiderability for the on site SEO, but a really well written guide much better than most 500 words articles that don't say anything! Great stuff!

Greg Gillespie said...

Nice post Tony, a great intro for website owners trying to understand the value seo presents to their businesses.

Only one problem though with your article, the thing you don't point out is what the "farming" startup is focused on selling.

If it was "farm machinery" then your article and subsequent conclusions would have been different. Also the startup selling farm machinery would definitely not be trying to rank for "farm machinery".



If I am looking for a "combine harvester new or used", I am looking for something in particular. Which incidentally is how people search in the real world when they are focused on buying, something in particular.

The job of an seo consultant or internet marketing professional is to assist their customers by clarifying what are worthwhile keywords to pursue. There is no point is spending a long time and a lot of money to rank highly for a keyword that will rarely convert to a sale.

Your article goes a long way to demonstrating this but falls short on this one critical point.

Cheers From Down Under
Greg Gillespie
CEO of HeliumSEO

Tony Karrer said...

@Surf - good call that part of what any startup should be doing is setting things up so they are good from a user and spider standpoint.

@Greg - I'm not sure I agree that the analysis would be all that different. Maybe I'm missing it, but wouldn't you still go through and figure out what and how much traffic you could capture if the specific focus was selling particular kinds of harvesters.

Possibly you are saying that before they dismiss SEO, the founder should spend more time doing deep dive to figure out how people are searching and if they can capture that. I would agree.

BTW, I agree that you will probably be targeting specific kinds of keywords as you describe for your example. And it's possible that the value of a click in that particular situation is high enough that optimizing for relatively little search volume is worth the time. The volumes in your particular scenario are going to be small.

Alon said...

Hi Tony,

you're absolutely right - in terms of Google juice not so much.

What I meant was more to the tune of: if i start "ABC company" today and submit my articles/pr's with links about it to a couple of the good directories, I can be certain that googling "ABC Company" next week will result in a couple of my articles/pr's showing up on the front page of google. I've now instantly controlled what a user sees on the front page when they google my startup!

Good initial base of first page results...and from there on it's onto more quality methods of SEO, as you've described.

Porsche Nguyen said...

Tony, nice post. SEO is an important online marketing strategy, but it takes skill and time before it makes much of an impact (as you speak to in your article). Creating a deep site loaded with quality content, publishing geo-specific "long tail content", generating a high number of relevant inbound links - these are all elements of SEO that work against any fresh new URL.

One of the reasons we created is to help companies (both start-ups and established firms) measure how effective they are at making their brand visible online. Search marketing is a big part of our Heardable Score...SEO for CEO's as we like to refer to it.

Your readers might want to try it out.

Best, - Porsche

Schikowski said...

Hi Tony,

Great post and a very important topic.

What I noticed is that you may want to re-think the numbers you provided for click-through-rates on positions 1-10. I assume you are referring to organic search results. From my experience, and according to numbers I've seen, #1 gets as much as 65%, #2 and #3 a combined 20% and the rest isn't even worth mentioning. The traffic you can expect from a #1 ranking is so much higher than what to expect from #2, that it isn't practical to start with targeting search terms where you won't have a chance to rank #1.

On the other hand, your search terms are what they are, and so is your competition. That's why I like your approach of targeting the long tail first.

The thing is that in the long run you want to target high volume search terms to really start getting traffic. So I would target one long tail search term after the other, and try to grow the site to take on the "big guys" at last.

I think your strategy provides a great starting point and can easily be adapted by startups.

Tony Karrer said...

@Schikowski - thanks for the comment.

I used a couple of different sources for the click through rates on different placements.

From my experience, the rates are highly variable based on the kind of search being done.

I should point out that the numbers I'm showing are actual clicks - which is about half of the searches. So, if I take your numbers 65, 20, 20 and divide in half:

32, 10, 10 - not all that far away from mine.

And like you said, the drop off is pretty steep.

I believe we are pretty much on the same page, except for your last statement:

"can easily be adapted by startups"

I think many startups interpret that to mean that they can easily drive enough traffic to be meaningful. You need to be careful on that assumption.

SEO Stratagem said...

Is the website Good?
This might sounds like a funny thing to say as a guideline, but I think it bears mentioning. An economical SEO firm can’t do SEO for a website that comprises just a few pages of flash, or a few pictures embedded into 3-4 pages. SEO is a major commitment. Before you start SEO, you need to have a professional looking site on a stable server. Without this, the economical SEO firm is not going to give you the results you desire.
From now on, I think that traditional SEO will get into the top 10 on Google. But, to get to #1 for a popular search phrase, you will need to painstakingly keep improving and editing your website until you have really low bounce rates. My SEO guideline is that you need to make sure that when people get to your website, they find what they’re looking for. I reached this conclusion after reading about personalized search. I think personalized search will be of great concern to many SEO companies. Personally, I think “personalized search” is the holy grail of algorithms to rank websites. It’s virtually impossible to fool and will continue to change the SEO industry.
If you're looking for a good introduction to SEO guideline, check this one out. It was written by google, and it's really good.

Schikowski said...

Hi Tony,

Targeting long tail phrases is a good starting point for them to experiment with SEO because they can see results fairly quickly. I like to advise newbies to target an easy search term until they achieve a #1 ranking, then target the next one and so on, until they have enough experience to target major keywords.

Whether or not this would drive enough meaningful traffic depends entirely on the business they're in and on the search volume of their keywords.

That's what I like about your post, you are giving them the right point to start out.


Tony Karrer said...

@Schikowski Great point. If you pick a few long tail term that has enough volume and work on those only, you can get a sense if the numbers work out.

You can probably do that pretty easily, but you definitely won't get the payback for the initial amount of time. But it would feed nicely into your analysis process.

Good point!

aysh tulip said...

hey u have done great work Mr. i also agree with Alon's point.
keep it up n best of luck for future life