Calendar
Network
Feeds
- 60
- Sphinn It!
Went Hot: March 25, 2008 - 9:06 am
Posted By: DoshDosh 47 days ago
Topic Type: News Story (Jump to http://www.seomoz.org)
Category: SEM Industry
12 Comments
Went Hot: March 25, 2008 - 9:06 am
Posted By: DoshDosh 47 days ago
Topic Type: News Story (Jump to http://www.seomoz.org)
Category: SEM Industry
12 Comments
Top 10 Most Sphunn
In What's New
- Tips For Dealing With Information O... (21)
- Confessions and Reflections of a Fo... (17)
- Live Twittering A Conference? Try L... (17)
- Comcast + Twitter = Perfect Feedbac... (15)
- Sphinn is a Dysfunctional Social Ne... (14)
- Five Tools Everyone Working Online ... (14)
- Need to hire SEM help? Where to fin... (14)
- SEO for Blogs: If You Build It, Wil... (11)
- What Every Good Marketer Knows (10)
- Social Media Influence: The YouTube... (10)
Latest Comments
- Success through simplicity... Seth certainly has a way bringing everything …
- Thanks, seobro, but I don't think I actually had any quotes regarding conte…
- Great work Jill as always. I have to give you credit on convincing people a…
- Good post, but I think the comments relating to the submitter are a little …
- Thanks for the heads up!:)
- LOL talk about lazy SEO!…
- Excellent post and something to save for the next time I deal with a design…
- I sometimes take issue with the way Seth Godin comes across but I agreed ab…
- - read more -
Save the date for:
• SMX Madrid (in Spanish, May 20-21)
• SMX Advanced - Seattle, WA (June 3-4) Register today! Early bird rate expires May 9!
• SMX Local & Mobile - San Francisco, CA (July 24-25) (July 24-25) Pre-agenda rate expires May 2. Get the lowest rate by registering now.
• SMX East - NYC - (Oct. 6-8)
• SMX London - November 4 & 5, 2008
Comments
An excellent insight into our industry. I'm quite surprised by the high number of people earning under $60k still - is this skewed by PPC employees (who tend to earn much less than SEOs in the UK until they become Head of Search)?
Rob - I think there are a couple of key things to remember when looking at the earnings data.
- People working in foreign countries where money is worth more
- People who are 'in seo' but in reality are small site owners or bloggers who don't really earn that much from the web,
- Yes probably PPC employees and other monkeys. Breaking £30k in the UK is relatively difficult outside of London unless you're head of search or similar.
(disclaimer, not that I'm saying all PPC employees are monkeys. That's not what i meant at all. Err..)
Yeah, also what about folks who work but also have their "own thing" on the side?
A limited survey labeled as representative of an industry. Sorry Rand, but calling it "the SEO Industry Survey..." and then disclaiming with this :
makes it not an SEO industry survey at all.
so there are people out there who read 400 work emails a day, and people who read over 100 rss feeds each day. When do they work?
I was pretty shocked to see that 55% of people think Google Analytics is a factor in search rankings. 31% think adsense ads are a factor? For serious?
PPC is not SEO, at least not anywhere that I know of. Maybe this should have been called The SEM Industry Survey Results? I know opinions vary but to me PPC (or paid search in general) and SEO are subsets of SEM. SEM implies knowledge/use of both.
Overall there's some interesting info, just surprised to see PPC lumped in as SEO.
@ darkmatter, maybe it's all the PPC folks skewing those results? PPC folks are not SEOs and a lot of PPC pros I know have little if any knowledge of how things work on the "other side of the fence".
"I was pretty shocked to see that 55% of people think Google Analytics is a factor in search rankings."
Wait till you see what I've got on GA :).
@ John: I don't get the criticism. Is your point that it's largely about PPC [which is not SEO]? Seems like a trivial quibble, since the naming could just be the sem industry survey. And fyi, I'm an SEO and ppc guy, and I answered.
@ Gab: I read John's criticism to be more about the number of respondents.
Ohhhh I get it. So the sample size is not 100%. When is it ever? 2000 (or 3000 for some Qs) is a great sample,imho. It's not like he's saying it's representative of the US pop.
@Gab: Sorry if I didn't make myself clear. I come from a background in experimentalism, which includes "Design of Experiments" and (obviously) statistics.
There are reasons companies like Gallup get millions for conducting surveys, and small Universities like Quinnipiac are quoted at every election. It is SIMPLE to make a survey that looks good, and VERY HARD to make one that is meaningful. This one "looks good".
Sample size doesn't need to be "100%". In fact, it needs to be exactly some number (very,very much less than 100% by definition) to enable you to make a claim. Mathematics determines how many people you need to ask, in order to give the answer you want to give.
So you might need a mere 500 people to predict the outcome of an election, within 2%. That's math, and it works. You've heard the old "there are liars, damn liars, and statistitians" because people don't trust statistics. BUT, again by definition, statistics don't lie. So where is the wiggle room when doing a survey of any kind (or an experiment, where you make measurements)?
The wiggle room is in the assumptions made (and not mentioned), and the bias that exists in the sample you tested.
Now you know how to critique any experimental report. Jump to the selection process (how they chose the participants) and check what was assumed going into the study.
This study is called "SEO Industry Survey" yet the sample (people questioned) do not represent the SEO industry well enough to satisfy the mathematics, and thus the statistics. If you choose to collect participants via seomoz (perfectly valid) then you have to characterize them for their ability to represent SEO (not done) and then let the math tell you how many you will need to ask (and will need to participate) in order for you to provide data (results) which are, in fact, representative of the SEO Industry.
The way this was done does not reflect on the SEO industry for many, many many reasons. Rand can make his surveys and they can be very meaningful to some people -- no criticism there. But when he labels it "SEO Industry Salary Survey" that, in my ever so humble opinion, colors it black. If no one says anything about that, the general public has no choice but to accept that it is accepted (within the community) as valid.
See how assuming the pedastal as a figure head carries inherent responsibility? And therefore, participating in said community carries inherent responsibility as well... to not speak out on an issue unless you are asking a question, or you know something about that of which you speak.
We shall collectively determine the importance of Sphinn as a community, with our public actions (who Sphunn, who DeSphunn, Who commented, etc). Similarly we collectively determine the effectiveness (and viability) of SEO as a career, via our public actions. So far it's not going so well, but I still have hope.
Rand hired an outside firm to conduct this survey. I'm sure he'll be able to defend the methodology with ease. He did an in-depth critique of Loren's Search Engine Journal informal survey.
There's no way to judge the survey accuracy without raw data.
John, that's a really in-depth answer. Nice to get more detail on what you meant.
All that said, I think that a survey of 2000 SEOs/SEMs is a fairly large sample. As you mention yourself - 500 to predict the us election outcome.
But your point isn't about quantity, but rather quality - i.e. representativeness. And it's skewed to SEOmoz readers. YOur point is well taken.
That being said, the survey still has value for representing a large chunk of the industry. Perhaps not the whole biz in statistically representative manner ... but certainly a wide swath of it. I know SEOs who don't read SEOmoz (just met a few this week actually). N some of them do OK. But I read regularly (religiously even - there u go, I'm biased like the sample ;) ) and outrank em :D hehe. So perhaps it's the survey of quality SEO professionals ;)
p.s. Awesome to learn that you're also a badass stats pro. I learned some in college in research methods and quantitative methods, which was cool. Nowhere near your level though!