21 replies to this topic

[iamlost]

In dazzlindonna's thread Social Media Shares As A Metric For Panda? I said I am more determined than ever to move traffic percentage away from SEs. To which EGOL replied I agree, but that is a really tough job. then asked What methods can we use to accomplish this? We are already working on good content.

As is often the case in webdev there is no one answer or magic solution. Especially as the real needed/wanted/desired solution is qualified traffic. In this context qualified traffic == traffic more likely to convert (define conversion as you will). Given EGOL's foundation of good content we are left with identifying, analysing, and optimising for potential traffic sources. And then marketing appropriately for each in turn.

This is not a step by step how-to (I have invested a lot of time and effort and even some cash into most of the following suggestions), rather a list of researchable actionable suggestions. The first thing to do is to thoroughly understand your business model and unique selling proposition(s)/point(s), market and market segmentation (behavioral, demographical, geographical, psychographical). Yes, know thyself.

Then decide how to differentiate traffic source and to what degree. Such a breakout might look something like:
* direct: type-in, bookmark.
Note: many analytics programs default classify any traffic for which they can not determine a source as direct traffic.
Note: some analytics programs default classify ppc traffic as direct traffic.

* search: Bing, Google, Yahoo et al

* directories: Yahoo, BOTW, JoeAnt, YellowPages, Yelp et al
Note: there are both free and paid directories/guides.
Note: which directories are best is vertical/niche/site dependent.

* social: MySpace, FaceBook, Twitter et al

* PPC (Pay Per Click), i.e. paid advertising: AdWords, adCenter, YellowPages et al
Note: I have never used PPC. There is a steep learning curve to do it right but the ROI can be significant. It can also be throwing money away.

* publish-republish elsewhere: guest blog, ezines, YouTube et al
Note: Many such third party sites were greatly impacted by Panda so care in selection is warranted.

* OPS (Other Peoples' Sites): last but not least the backbone of the web, all those delightfully discerning sites that link out to yours.

Let's walk back up that breakout in reverse order:
* OPS: here is where the foundation of 'good content' is perhaps most important. Why should another site link to one of your pages? There are two major link/traffic building approaches:
---build subject link magnets (the upscale version of link bait) optimised to of captivating interest and gold plated resource to your niche audience.
---research specific desirable sites and build subject matter that extends or compliments their offering(s).
Note: how you then offer/market your creations is your business.

* publish-republish elsewhere: there can be a significant difference in approach and offering depending on your site focus, i.e. personal service, ecommerce, information. As my experience lately has all been with an information focus I have retained most content for on site use only. I do, however, utilise YouTube (video), Flickr (images), Amazon (ebooks), Slideshare (presentations). I also appreciate Q&A sites such as Yahoo Answers and Quora, contributing to Wikipedia - much less time and attention required than 'standard' social media but with some of the aspects.

* PPC: not my thing. caveat emptor.

* social: the real trick is NOT to follow your competition but to follow your ((potential)(most) converting) visitors. And the kicker is to only go where and how you can be of value to those visitors. With social the how you engage/converse is critical. Have both strategic and tactical plans for each platform/application (they are each quite different) before you start and be willing and able to adjust as necessary. And be ready for the long haul, jumping in and then out is a waste of resources at best, a damage to your reputation at worst. It is far far better to have not engaged than to have done so poorly.

* directories: have seen the long decline from glory days of old reverse with the advent of Local search. A majority are still not worth utilising for a number of reasons, however, careful selection and use can be quite advantageous.
Note: I lump what some term 'guides' in with directories.

* search: not much to say here except that diversifying/expanding search traffic from one to several search engines is worth exploring.

* direct: the possibilities may vary somewhat depending on site and business model. One universal is to work at retaining visitors both to stay longer and to return, primarily through content design and architecture. A second is to build a solid, preferably remarkable, brand; when visitors think of your niche they should think of you.
Some direct traffic generators worth considering:
---as EGOL mentioned, RSS.
---email newsletters, aka mailing list
---downloads, i.e whitepapers, graphics/videos, apps, widgets...
---a blog to encourage conversation
---suggest/remind visitors to bookmark pages.

[A.N.Onym]

There are plenty of opportunities for off-search traffic, as you've mentioned, but my two top priorities would be:
- cornerstone top quality content on my own website, "subject link magnets", as you call them, to attract links, search and referral traffic, which would, in turn, boost..links, search and referral traffic, which, of course, would..
- high quality articles on very related websites - blogs, journals, etc in general or related fields, where I can comment on the matter from my own angle (e.g., weight loss with running, rather than only dieting/other stuff). This can be offered in downloads, but I'm not going to create a separate downloads section just because I have to have PDFs and other content won't be in PDFs.

Republishing my own content would be wasteful to my future traffic, unless it's microcontent, which I never counted on to get major search traffic and links to begin with (tweets on Twitter and republished elsewhere, video - both on my site and on Youtube, photos - both on my own site and on Flickr, etc). It doesn't mean that videos can't bring traffic, but they should accompany or be accompanied by plenty of top quality content on the same page to be the volumes I need, imho. Unless it's breaking news or something.

With that in mind, I will probably omit:
- on-the-beaten-path directories, since everyone has done them, they've lost most of their value. I might pay attention to selective editor-based link collections on topical websites, but not general-topic directories
- article directories: the value of them is almost nill, and the time to write/publish the articles is huge, comparing to cornerstone content and social promotion (and even guest authoring) - unless it's cheap (re)writing and automatic submission. Still, money is better spent elsewhere.
- off-topic linkbait: good to build links, but I'd rather stay on topic, even if it means making a better effort
- blog directories (who subscribes from them these days?)

Note that we've been talking about the usual stuff here. But to get off-search traffic, we have to build an off-search platform. It has to be one or more of these:
- an online web service or two (ideally, around local, mobile, physical goods, sharing, social meeting/responsibility, dating among hobbyists/professionals, if you will)
- a community around the matter to discuss and do stuff
- a shop
- a platform to share physical things (bartering, selling, etc)
- videos (including video reporting)
- audios

Basically, it has to be more offline, than online, to get off-search word of mouth traffic. Thus, I think it's long been necessary to come full circle back to offline and bring it online. After it's all done, it'll just be a normal website

Edited by A.N.Onym, 10 July 2011 - 05:04 AM.

[DCrx]

I question whether people are developing good content. What they overwhelmingly gravitate to is "good enough" content.

I question whether you can even entertain the concept of wholism with SEO. It's still a traffic discussion. Once you start writing for the 'bot, you don't just stop.

Whan a killer app for this topic? Start thinking about the user's ability to bookmark your site ...and then, return.

Edited by DCrx, 10 July 2011 - 06:13 AM.

[EGOL]

Thank you for starting this great thread and doing that with such a detailed presentation!

I think that "social" has a lot of potential but as you point out they are not all the same. I tend to place social sites into two categories for building traffic.

Sites that communicate to your followers: Sites such as twitter and facebook can be effective at getting traffic if a large number of well-followed and influential people are following you. If you can share great content that appeals to them it will be passed along quickly.

Sites that have their own following: Sites such as reddit, slashdot and digg are very different. People arrive at those sites looking for something interesting to read. They want to find great content. If your content becomes popular on one of these sites you will get a lot of attention. These sites are great if you are just starting out or if you don't have a lot of followers. An article that becomes popular on one of these sites can very quickly pull many thousands of visitors and those visitors will often share it with their own followers and you will have traffic coming into your site from a large number of directions.

[DonnaFontenot]

Unless you are in a market that has little to no search volume, even for the number one sites, or you get zero to very few visits from search because you just don't rank well (or you've blocked bots), there's little hope of not becoming reliant on search.

[sounds of screeching tires erupt at this point, as people prepare to argue] :hmmmm:

The only way to not become dependent upon search, in my opinion, is to choose to not receive visitors from search.

If you get no traffic from search, then no dependency develops. You are free to become reliant on the traffic you receive from all other methods. Let's say an online business sells bananas. The business gets no traffic from search. The business has done a good job of diversifying via all the means posted above. While there is likely to be some ups and downs in the volume from day to day and month to month, it will still remain at a mostly predictable level. So the business can determine how many banana trees must be maintained on how many banana plantations, and knows how many banana pickers must be employed, and how many packers, and how many shippers, etc.

The second that search becomes involved, however, is the second that the banana business becomes dependent upon search. If bananabiz.com suddenly starts getting flooded with visitors from search, bananabiz must now ratchet everything up. More trees, more plantations, more pickers, more packers, more shippers. Bananabiz now has 100x as many people reliant upon the biz receiving the traffic it receives. It hums along nicely that way for a few years.

:nanaparty:

Then, BAM! Something like Panda arrives, and all that search traffic goes away. Mr. Bananaman cries in his bananaless cereal, as he lays off tons of people and watches tons of bananas rot in the fields. Suddenly, a million braggarts are telling Mr. Bananaman that he was a fool to rely on search traffic.

:nanaontheside:

I say that Mr. Bananaman had no choice but to rely on that search traffic, unless he made the decision from the beginning to block his site from the bots. As long as the possibility existed that he would receive tons of traffic from one source, the possibility of becoming reliant on that traffic was inevitable.

Of course, that scenario is for ecommerce, but it works the same way for information sites. X amount of traffic = X amount of income = X amount of expenses. Lose the traffic = lose the income = cry in bananaless cereal.

Now, don't get me wrong. I love getting traffic from diverse sources, but I firmly believe that what I've stated above is primarily true, with possibly some edge cases that might prove otherwise (such as if Mr. bananaman has lots of money saved up that he can use to keep things going while he attempts to gain the search traffic back, or finds a way to replace it elsewhere). And really, when I say "search" traffic, what I really mean is ANY ONE SOURCE of traffic. Whatever that might be. If one source brings in tons of traffic, and then dries up, the same applies. The only way to not become dependent upon any one source of traffic, is to not get much traffic from them to begin with.

[DCrx]

Unless you are in a market that has little to no search volume, even for the number one sites, or you get zero to very few visits from search because you just don't rank well (or you've blocked bots), there's little hope of not becoming reliant on search.

Any diehard SEO or their myriad supporters, argues this is when you hire an SEO firm. I'll leave the arguement of amping up traffic at the expense of targeted traffic for them to make.

The only way to not become dependent upon search, in my opinion, is to choose to not receive visitors from search.

This is okay, but isn't going to be popular. It'd be nice if every site blocked bots, concentrated on improvement and conversions, then only went to search engines from a position of strength. What happens is, they get 0.0025% conversions; then the SEO says the only way to improve sales is with 5,000 times the amount of raw, unfiltered, disinterested traffic.

And probably a dedicated server and premium bandwidth package. My guess is there's a finger in that pie as well -- much like bloatcode aids hardware sales.

[DonnaFontenot]

Hmm, I think you've completely missed my point, but that's ok.

[DCrx]

I think I got the point. Just that there's an SEO dependent answer for it.

Imagine a site that did the unthinkable and targeted (gasp) their customer base. I mean built a site, just about exclusively as a bonus service for customers. Practically zero outside exposure whatsoever.

All backend. Almost zero front end. In essence the opposite of the front end loaded Flash splash page site.

Imagine that site had the goal of upping customer lifetime value. (Remember when there used to be customers?)

Edited by DCrx, 10 July 2011 - 09:23 AM.

[EGOL]

I would like to see traffic data, broken down by source, for some of the large popular sites.

My sites seem to get at lease 80% of their traffic from search. I get a lot of traffic from other sources but search just overwhelms it.

Any site that has the ability to pull a lot of non-search traffic will probably rank well in search because people are linking to it (and doing whatever else search engines monitor to identify sites that people visit and use). Search traffic is a byproduct of the ways that people use and share a site that they enjoy. And, sometimes a byproduct can become the primary product by volume even if that was not by design.

Getting to the bottom line... Would you agree that high quality content and tools to make it easily sharable are the keys to getting no-search traffic?

Hmm, I think you've completely missed my point, but that's ok.

Then I wouldn't have gotten a chuckle out of....

...much like bloatcode aids hardware sales.

[Ron Carnell]

I agree completely with Donna, but would come at it from a slightly different perspective.

Everyone talks about being visitor-centric. Even DCrx, although he apparently thinks everyone follows a customer based business model. So, what happens when we look at traffic acquisition from the visitor's point of view?

What do surfer's currently do when they want to find something on the Internet?

I submit to you that their first step is always going to be a search engine. Only if they fail to find what they were looking for will they widen their parameters to directories, social media, or random links on Other People's Sites.

I think most of the items in iamlost's list are worth exploring (and would, personally, add community to the top of the list), but I just can't see any of them, or even all of them, replacing search engine acquisition any time in the near future. Not if one hopes to consistently attract more than a few score visitors a day (which may well be enough for some business models).

iamlost said: Note: I have never used PPC. There is a steep learning curve to do it right but the ROI can be significant. It can also be throwing money away.

The difference between good and not-so-good PPC is exactly the same difference between good and not-so-good organic SEO. Keyword research (and constant testing) is the key to success for both. Done correctly, both PPC and SEO deliver highly targeted traffic. They deliver people who are actively looking for whatever it is you have to offer. It just doesn't get any better than that.

Search engine traffic is like putting up a big sign on the road for your newly opened grocery store. Assuming your sign is accurate, you can be fairly certain that most of the people subsequently walking through your door are there to buy groceries. You still have to do a good job once you get them in the store, but that's another issue entirely.

Directories, social media, and random links? Those are more like the little sundries you might put in the aisle leading to the cash register. They're not often the things people are actively looking for at the moment they seem them, but you can reasonably expect to snag a few impulse buyers. And that's a good thing, too.

I'm reminded of the old story about the slightly intoxicated fellow who was down on his hands and knees beneath a street light looking for his lost car keys. He knew that wasn't where he lost the keys, of course. But the light was better.

I would suggest you don't want to spend too much time looking for traffic under the street light. It ain't there. For most of the past ten years, it's been on the far side of the road, in that very dark, more than a little scary ditch . . . called Google. You don't have to like that, of course.

You still have to get in the ditch, though, if you want to find your keys.

[iamlost]

Unless you are in a market that has little to no search volume, even for the number one sites, or you get zero to very few visits from search because you just don't rank well (or you've blocked bots), there's little hope of not becoming reliant on search.
...
The only way to not become dependent upon search, in my opinion, is to choose to not receive visitors from search.
...
The second that search becomes involved, however, is the second that the banana business becomes dependent upon search.

What do surfer's currently do when they want to find something on the Internet?

I submit to you that their first step is always going to be a search engine. Only if they fail to find what they were looking for will they widen their parameters to directories, social media, or random links on Other People's Sites.

I agree with you both.
Ha!
Bet you didn't expect that

However...

while search is the 'easy' 'simple' default traffic generator it is a business model decision, a mindset, that says let's stay there (with search, usually Google) and not actively build out other sources.

I continually hear from Google-centric webdevs that while Bing and Yahoo may (niche/site dependent) convert significantly better they supply such little traffic not worth pursuing... do you see the contradiction in that?

I view search as the current web traffic default, much as AdSense is the ad revenue default and Commission Junction is a common affiliate default. And one can, of course do well, with defaults. However, defaults have two main drawbacks:
* all revenue eggs in one (or two) basket
* probable less revenue share/conversion percentage

Is it simple or easy to diversify?
Either traffic or revenue?
In meaningful amounts?
Usually not.
I've been working at it for years. And while few attempts crashed and burned many did just sputter and fizzle.

The point that is missed by many is that when I recommend traffic diversification I DO NOT mean lessening search traffic but rather maintaining search numbers while lowering it's percentage overall.

My sites seem to get at lease 80% of their traffic from search. I get a lot of traffic from other sources but search just overwhelms it.

What I do is track what I call gross and net traffic from each source where gross==total and net==converting. It is important to know the value of traffic and it's source.

For arguments sake I'm going to imagine a few absolute numbers to go with your percentages.
100% = 1-million uniques per month. Therefore search provides 800,000 and all others 200,000.
The traffic diversification goal should be to maintain (at least) the 800,000 from search and while working to grow the other. After 6-months you are at 1.3 million with 900,000 from search (up 100,000) or but down to under 70% of total with other doubled to 400,000 actual visitors and up 50% in percentage terms to 30% of total traffic.

If you also track conversion percentage for each traffic breakout source then you can target areas in order of probable ROI benefit.

The only way to not become dependent upon search, in my opinion, is to choose to not receive visitors from search.

I respectfully disagree.
That does not mean that search does not remain a major even dominant traffic source, just that one can become increasingly less dependent. My search traffic seems to be sticking at about 40% (all sites averaged), it's been up and down a few percentage points of that mark for the past couple of years, with Google fluctuating under 25% total, 62.5% of search, traffic.

So search is still my dominant traffic source, indeed Google is still the single largest referer, but dependency? No. And if I can do it, so can others.

With the increasing advent of social it should become easier for many niches to diversify from search Google. A matter of mindset, of going where your customers are and offering value such that they will follow you home and return to visit, with increasing frequency in increasing numbers.

I would suggest you don't want to spend too much time looking for traffic under the street light. It ain't there. For most of the past ten years, it's been on the far side of the road, in that very dark, more than a little scary ditch . . . called Google. You don't have to like that, of course.

You still have to get in the ditch, though, if you want to find your keys.

I like your analogy, Ron.
However I would turn your story on it's head and say that the street light under which everyone is looking is Google and it is the dark and the scary ditch where traffic other than Google exists.

[EGOL]

The idea of being "dependent upon search" has different meanings to different people.

If you are an affiliate or an ad publisher your risk is on the low side. However, if you are a retailer you might have a huge investment in inventory, a very expensive lease to house that much stuff and employees to run the warehouse and fill the orders. Getting bit by panda will hurt both of them but the retailer is going to be stuck with an awful lot of widgets and possibly a few years of lease payments to make.

I wonder if any of the insurance companies are selling panda insurance?

[tam]

I just checked my sites, the lowest search portion is 65%, most are 70-80%.

I agree, if I want to find something 9/10 I'm going to google. If I follow a link on someone else's site, I've probably already googled to get to that site

I do wonder about diversifying search traffic though. If 90% of SE traffic comes from google, then anything that effects google ranking has a big impact. If the traffic was spread more evenly between search engines then any change on an individual SE would have less of an effect. Anyone know what the portion of users are between the top few SE's?

[DonnaFontenot]

Compete puts out a search engine market share each month. Here's the May one:

http://blog.compete....t-share-report/

And here's a tiny portion of that chart:

 mayshare.png   2.35K   33 downloads

Basically, it's a 2/3 to 1/3 breakout of Google vs. Bing (and Bing is "bing-powered", so that includes Yahoo within it).

Everyone else, ask, etc. is practically non-existent.

[iamlost]

I have issues with Compete and Quantcast methodology but that said my question to webdevs is: how many of you are enjoying that one third of search traffic that Bing/Yahoo serve?

From prior threads here and elsewhere Bing/Yahoo site traffic is often non-existent to <10%. That is a major traffic miss - depending on niche it could be a critical miss (the three search engine demographics overlap somewhat but do show significant differences).

[DonnaFontenot]

Right, iamlost. If the niche caters to old people, or very un-techsavvy people (like AOL types), then Bing/Yahoo should be something to concentrate on. Good traffic and good conversions from those folks if your site appeals to that crowd.

[bwelford]

Perhaps we need to back up a little here. I do not believe all queries are necessarily of equal value. It is interesting that the searches per unique visitor in those Compete numbers show that Google searchers seem to do many more queries than the other searchers. Here are the average numbers of queries per unique visitor for May 2011:

Google 69
MSFT 27
Yahoo! 27
Ask 6
AOL 14

I'm sure many of us have got used to typing in the name of a website rather than the URL. So it may be that there is a 'base-line' number of queries that just reflect that habit of using a search engine to find the website you know you want.

The queries that are of value are those where someone is looking for a source of a product or service without having a known website in mind. It would be great to have the figures for these types of queries. That might show a very different set of results for 'Market Share'.

[iamlost]

If the niche caters to old people, or very un-techsavvy people (like AOL types), then Bing/Yahoo should be something to concentrate on.

A rather broad brush stroke with an amount of truth.

Note: Take all following numbers with a grain of salt. However, as a relative comparison between SEs they are reasonably reliable.
Note: the base comparison below is with general US demographics, not directly with the other SEs (perhaps another time - although some inferences can be drawn).

1. A deeper US demographic breakout for Bing

Compared to average US demographics Bing is:
* Gender:
---very slightly above average in male users.
---slightly below average in female users.

* Race:
---significantly below average in Caucasian users.
---over 50% above average in African American users.
---above average in Asian users.
---over 40% above average in Hispanic users.

* Age:
---above average in <17 users
---significantly above average in 18-34 users.
---below average in 35-49 users.
---significantly below average in 50+ users.

* Children:
---above average in with children.
---below average in without children.

* Annual Income:
---significantly below average in 0-30,000$
---below average in 30-60,000$
---slightly above average in 60-100,000$
---above average in 100,000+$

* Education:
---significant above average in no college.
---significant below average in college.
---below average in Graduate School.

Note: Bing has an high return user return rate.

..................................................................................
..................................................................................

2. A deeper US demographic breakout for Yahoo

Compared to average US demographics Yahoo is:
* Gender:
---slightly below average in male users.
---slightly above average in female users.

* Race:
---significantly below average in Caucasian users.
---over 30% above average in African American users.
---over 40% above average in Asian users.
---over 15% above average in Hispanic users.

* Age:
---significantly below average in <17 users
---significantly above average in 18-34 users.
---above average in 35-49 users.
---significantly below average in 50+ users.

* Children:
---slightly below average in with children.
---very slightly above average in without children.

* Annual Income:
---below average in 0-30,000$
---below average in 30-60,000$
---average in 60-100,000$
---above average in 100,000+$

* Education:
---average in no college.
---average in college.
---slightly above average in Graduate School.

Note: Yahoo has an extremely high return user return rate.

....................................................................................
....................................................................................

1. A deeper US demographic breakout for Google

Compared to average US demographics Google is:
* Gender:
---very slightly above average in male users.
---slightly below average in female users.

* Race:
---slightly below average in Caucasian users.
---slightly above average in African American users.
---over 20% above average in Asian users.
---over 20% above average in Hispanic users.

* Age:
---significantly above average in <17 users
---significantly above average in 18-34 users.
---below average in 35-49 users.
---significantly below average in 50+ users.

* Children:
---above average in with children.
---below average in without children.

* Annual Income:
---significantly below average in 0-30,000$
---below average in 30-60,000$
---slightly above average in 60-100,000$
---above average in 100,000+$

* Education:
---slightly below average in no college.
---average in college.
---above average in Graduate School.

Note: Google has a very high return user return rate.

The above are very basic demographice breakouts. Webdevs serious about such metrics dive much deeper into not only SE traffic but that for all significant referers. Statistics, yum

[iamlost]

Perhaps we need to back up a little here. I do not believe all queries are necessarily of equal value. It is interesting that the searches per unique visitor in those Compete numbers show that Google searchers seem to do many more queries than the other searchers.

Barry

I am much less interested in number of queries - the most commonly reiterated statistic - but in the number of unique visitors. And that turns things very much on their heads. For one thing it removes all those automated Google aimed SEO tool queries...

Using Donna's Compete May SE stats link:
Google: 138 million
Bing: 93 million
Yahoo: 81 million
Or Bing and Yahoo together saw more unique users in May (174 million) than Google (138 million).

As I keep saying: there is more to Search traffic than Google. Much more.

[bwelford]

I beg to differ here.

If all those Bing and Yahoo searchers were merely using their search box to go to their Facebook page, then those unique visitors can be ignored in considering which queries can usefully be targeted.