Tuesday, January 27, 2009

Two tools for tracking search keyword placement

I've been working over the past few months on better search-engine results for several key phrases related to KidPub. There are two tools that I'm using to help track them:


SheerSEO
This site monitors weekly changes in placement of search results on Google and Yahoo. It's currently free to sign up. You can track a large number of search phrases, and the results are shown in an easy to understand table. You can also export the results to a Excel file. Here's a snip of a recent report:




You can see that 'kids stories' is moving up nicely, from 37th spot to 24th. You can also see that 'publish my story' doesn't point to KidPub's home page, which I need to fix.

SheerSEO's service runs once per week and sends an email to you when new results are ready. You can add or remove search terms at any time.

You can see a position chart for each phrase on either Yahoo or Google:



I think that this is really useful information, and it saves me the trouble of having to track and chart a spreadsheet full of keywords. I can measure the effect of changes I make on KidPub, both positive and negative, and know when I can stop tweaking.

SEOBook Rank Checker
For spot checks I like to use SEOBook's Rank Checker, which is an add-on for Firefox. You can grab it from the SEOBook site. Even though it does offer historical information, I tend to use it when I want to know how a particular keyword is standing on Google or Yahoo (it also does Live Search). There's a scheduling option that will run the checker periodically for you in the background.

The results are typically identical to SheerSEO's.

Both tools give you a great bit of intel on how your SEO work is progressing, and you can quickly see if something has either gone wrong (fix it!) or is working well (duplicate it!). This level of detail might seem overwhelming, but remember, half to three-quarters of your site's traffic is probably coming from Google and Yahoo, so it make sense to pay attention to search reult positioning!

Monday, January 26, 2009

Closing in on 100K!

Back in August, KidPub was ranked around 400,000 by both Compete and Quantcast. We targeted the 100,000 mark back in November when we started doing serious work on building traffic, mainly through SEO and grass-roots marketing. I just looked at today's Quantcast number...119,146. Last week was 127K or so. We're getting closer!

Once we hit the 100K mark the hard work begins...gettting KidPub into the top 50K web sites. I know there are ways to 'spoof' Alexa to get a high ranking, but Quantcast numbers don't lie, since they are directly measured.

We're not getting many conversions yet from the Facebook ads, but the exposure is helping a bit, I think. It's only been three or four days since the FB campaigns started...it needs more time before I can really analyze the results. We're still watching click-through rates on several ads to figure out which are the best performers.

Friday, January 23, 2009

Free (Ad) Lunch on FaceBook

I'm not sure why it took me so long to figure out that Facebook is a seething pool of KidPub's target audience. The site is aimed toward kid ages 8 to 15, and there certainly are a few of them on FaceBook. A blog post over at StartupNation prompted me to take a look at advertising opportunities on FaceBook.

Ads on FB are similar to Google Adwords. They are small, textual (though you can add an image) and appear in an ad bar at the right-hand side of the page. If you run Adwords campaings you can likely run the same ads on FB.

What is great about FB advertising is that they offer you the option of targeting a specific audience, and provide a rough estimate of how many users are in the particular set:


My target audience is kids 15 and under, and FB estimates about 2,000,000 users in that range.

Like Adwords, you can set daily spend limits, set the bid for either CPM or CTC ad types, and track progress. The first ad I placed has run for about 12 hours with 142,000 impressions and 64 clicks. Not a great ratio, but my cost-per-click is lower that on the Google ad network, and I hope that I'm reaching a higher concentration of my target audience. I'll post about conversion rates when I have some data.

Ah, the free lunch part. The StartupNation blog pointed out the value of advertising on FaceBook. I started doing a bit of resarch, and kept running acroos bloggers saying, "For goodness sake, don't spend YOUR money on ads...use coupons!"

Sure enough, a bit of Googling turned up lists of coupon codes to redeem at FaceBook for advertising. I spent about half an hour hunting and submitting, and ended up with $350 credit in my ad account. Coupns are used before your primary source of funding (a credit or debit account), so for a week or so I can play with ad styles, formats, and copy and see what works well, using FaceBook's money.

Thursday, January 22, 2009

Quantcast Demographcs Improving Over Time

Not a huge surprise, but I've noticed that the demographics reported by Quantcast for KidPub have improved quite a bit. It makes sense; prior to adding their tracking code to my site they were making their best guess...now they can more closely correlate their data.

The biggest change is in the gender numbers. Based on my own look at registration data, KidPub has about a 60-40 girl-boy ratio. Before adding Quantcast tracking code they reported the opposite, or about a 58-42 boy-girl split. They now show the split to be 58-42 girl-boy, which matches my own numbers.

It helps build my confidence in Quantcast's numbers. Alexa and Compete don't provide this level of detail. And I'm starting to wonder about Alexa. The rank numbers given by Quantcast and Compete are nearly identical (Quantcast: 173,693, Compete: 172, 223), while Alexa is way off in the weeds. I've read that Alexa is relatively useless for site not ranked in the top 100,000, and I'm starting to believe it.

Sunday, January 18, 2009

Name recognition: HARO

Peter Shankman runs a great service called Help a Reporter Out (HARO). Two to three times on weekdays, he sends out a list of queries from reporters who are working on stories and looking for experts to interview or quote. As Peter points out, everyone is an expert at something!

HARO is a great way to build recognition and trust for your business and yourself as a brand. I experienced this while working as a market analyst for a firm in Boston. After one or two solid conversations with a member of the press, word got around that I knew something about enterprise IT infrastructure, and I started receiving regular calls when a journalist had a question.

HARO queries are all over the place topic-wise, so there's bound to be something that you know something about. By providing insightful commentary you'll start to build a reputation as a go-to person when journalists are writing, and the exposure for you and your comapny can be priceless.

Thursday, January 15, 2009

Alexa versus Quantcast versus Compete

"You can't manage what you don't measure" has been an adage as long as there's been management. It's also human nature to rank, compare, and contrast...I'm sure that the buzz around the water cooler in Egypt was, "So, are these the tallest pyramids, or have the Aztecs taken first again?"

I'm a big fan of Google Analytics and spend time every day looking for new ways to cut the data. There are insights in GA data that you just can't get anywhere else short of hiring a propeller-head to build some industrial-strength custom code for your site. The one thing that GA lacks, though, is competitive data, and for that I turn to sites like Compete, Alexa, and Quantcast. They all have strengths and weaknesses, and I use all of them for slightly different purposes. All three try to measure traffic and demographics on web sites and display them in a way that lets you measure performance.

Compete.com is the vaguest of the three in terms of where they get their data. According to their site, they use a "unique methodology" that relies on work done by "experts in the fields of mathematics, statistics and the data sciences." I have to admit, it reminds me of the end of Raider of the Lost Ark. They apparently have "Top. People." working on this. Kidding aside, Compete lets you compare five sites in terms of visitors, time spent on the site, and a few other parameters. I use it for a quick look at how KidPub is stacking up against my competitors. Data are presented in monthly chunks.

Here's how KidPub and Stone Soup compare at Compete.com:



Alexa relies on user participation to generate the bulk of its data. Web surfers must install the Alexa toolbar into their browser, which send daa back to Alexa for crunching. The advantage of this is a good degree of granularity...Alexa can say with a fair amount of confidence how long a user stayed on a site, how many pages the viewed, and so on. The huge drawback is that the data sample is self-selecting...people who a) know about the Alexa toolbar (maybe because I put a link to it on my site) and b) are willing to install it. I won't install it, and most tech-aware users won't, either. The data source is less vague that Compete's, but I still have to take a big grain of salt when evaluating numbers on Alexa, especially for sites that are not in the top 100,000 in terms of traffic. Still, I look at the data to see if there are nuggets to glean.

Quantcast is relatively new and is taking on Alexa and Compete head-on. Quantcast's big advantage is that they use both Google analytics style data gathering, where you need to install a snippet of Javascript on your pages, and correlation of any other bits of data they can find. For example, when a user comes from JoeSchmoe.com to KidPub, there's data in the HTTP header that indicates where the user was before they hit KidPub (assuming they came via a lnk), and Quantcast uses that to record traffic information about JoeSchmoe, even if they don't have the Javascript snippet installed. They also try to provide demographic information...age of user, income level, and so on...though personally I think that sort of data is a bit suspect. They also allow you to compare sites and have an interesting 'media planner' tool that lets you find sites that meet specific demographics.

Here's a Quantcast chart of demographics for KidPub:





All three are free and they complement each other in terms of what data you see. If I had to pick only one I'd go with Quantcast for the simple reason that I trust their data source. As you might expect, the reports are similar to Google Analytics...they should be, as they rely on similar data gathering techniques. I use Quantcast to track the 'rank' of Kidpub (we are, according to Quantcast, the 112,582nd most popular site on the web right now), and Compete to get a quick monthly look at competitive trends. Alexa is last on my list, and I don't really trust their data, but I do look at it weekly to validate my other sources.

Google Analytics tells me intimitate details about what's happening on my own site, and the other three give me hints about what my competititors are up to. All of it is valuable data.

Tuesday, January 13, 2009

Fun With Feedjit


Just a fun little post today as I work on getting a few books moved along the production cycle.

Several months ago I installed a little widget at KidPub from Feedjit. It's a map of visitors. "Oh boy," you're probably thinking, "Another map with little dots. How very Web 2.0."

But wait! Apart from the fun of watching your visitors move across the globe with the sun, you can actually gather some useful information from Feedjit. Clicking on the small map will take you to your Feedjit site, which is a Google Maps mashup showing each visitor and (here's the fun bit) what they're viewing on your site. It looks something lke this:



A little hard to see here (click for full-size), but you get stats on what's currently happening on the site, including the most popular page views in the displayed geographic region.

I don't use it that often, but KidPub members like it, and if we've published something specific to a region it's useful to take a look a few times during the day to see if interest has been piqued. Its a good way to take a quick look at traffic, too. The performance hit is minimal as the geolocation lookup is done on the Feedjit side. You are handing a third party data about your users, but hey, we all sold our visitors' souls out a long time ago, eh?

Monday, January 12, 2009

Google Analytics and Search: How Are People Finding Your SIte, and What Can You Do About It?

I've been focusing in recent posts about core keywords and working on SEO to get those keywords into better position in Google search results. There's another side of this story, though. It's great to have the #1 result for a keyword that you find relevant to your site. But what do 'real' users think is relevant? Sometimes you are too close to the business to think creatively about keywords.

First of all, if you aren't using Google Analytics on your site, shame on you! Go there right now and set it up. It's free, and provides a wealth of information about your site and your users, and you can't begin to improve things until you have some notion of what it is you're improving. Remember the adage: "You can't manage what you don't measure."

If you are running Drupal site, as is KidPub, there are modules such as Google Analytics that automate tracking of your site.

One of the things that GA tracks is inbound clicks, and you can learn a lot about how users find your site from this data, as well as what they do once they've found you. For example, I've spent a fair amount of effort getting 'publishing for kids' into a strong position on Google. It's currently the #1 result.

When I take a look at inbound traffic, focusing on Google search referrals, 'publishing for kids' is high on the list, but not as high as you'd think. It's number 10 overall, and a surprising 83% of visitors hitting the site with that search phrase are repeat customers.

What's the top seach phrase for inbound traffic? It's 'new twilight book', second behind 'kidpub'.

When someone does a Google search for 'new twilight book', the first result returned is stephaniemeyers.com (she wrote the series). But the SECOND result returned is a link to a review of the book by a KidPub member. Amazon.com and MTV.com are farther down the list!

Analysis of the behavior of these visitors reveals a lot of room for improvement. Ninety seven percent are new visitors...they've never seen my site before. They spend an average of 30 seconds on the site (the review is short, so that's plenty of time to read it). Then they leave...the bounce rate is 85%.

I look at this data and realize that I'm blowing a significant opportunity here. The experience at KidPub for these users needs to be different than for other users. At the very least, there needs to be links to other Twilight-related articles on the site. I've chosen three out of dozens and included them at the bottom of the review. I'll post the results when enough data have been gathered.

The point of all of this is that YOU might not see your site the way that OTHERS see it. By using tools such as Google Analytics you can gather valuable data about how people are using your site that will help you think about your site in new ways and to spot trends quickly enough to take advantage of them.

Sunday, January 11, 2009

Doubling website traffic with Drupal SEO optimization

[A cross-post from drupal.org] I've been playing with SEO optimization on my Drupal site (www.kidpub.com) since about August of last year, mainly tweaking meta tags (description) and title tags. I use Nodewords and Pathauto for high-level optimization.

KidPub is a site for kids who enjoy writing. It was launched in 1995 with custom code on the CGI framework, then migrated to PHP in 2000, and to Drupal in about 2005. We give kids a place to post their stories, poetry, reviews, and other creative writing; host writing contests with great prizes, and in general give kids a fun, safe place to hone their writing skills and build confidence. We also publish books written by children. I am one of only a handful of individuals who have run the same website continuously for 15 years, and I think it gives me some interesting perspective.

Traffic to KidPub in its Drupal incarnation averaged 3,000 to 4,000 unique visitors per month through August of 2008, when I began to look for ways to increase traffic to the site. About 65% of our traffic is generated from Google searches, 20% direct, with the remainder from Yahoo, MSN, and other search engines. It wasn't hard to decide to go after better Google placement. The goal was to double traffic by improving placement in the search engines in our target keywords.

The results are available graphically at my blog, http://kidpub.blogspot.com. In the period from August 2008 to December 2008, traffic has improved from 4,000 unique visitors per month to about 9,000 unique visitors, and KidPub.com is now the #1 search result on Google for 5 of our 10 target keywords ('kids publishing', 'kids stories', 'kids writing', 'publish my story', and 'kids publisher'). We are continuing to optimize to imrove the placement of our other 5 keywords.

Optimization was done in several steps. First, we used Pathauto to produce better search results by showing the Google spiders human-readable filenames based on the story title. A side effect of this is excellent placement on keywords that we hadn't thought of, such as 'new twilight book' (#3) and 'magazines that publish kids stories' (#4).

Next, we used Nodewords (Meta Tags) to set global keywords in meta tags for all pages returned at KidPub (currently about 14,000 indexed by Google). The keywords included our ten target words and phrases.

The title tag was modified using the page_title module to output a ten-word phrase that describes our site. Heat maps (graphical representations of what a user looks at) of Google searches show that users spend most of their short time on the Google page in the upper left, specifically on the links for the top three results. The link come from the title tag. It's your ten-word marketing opportunity, and it needs to tell the reader exactly what your site is. Our title tag emits 'Publisher of books and stories for kids, by kids'. You can see a heat map and an explanation of using the title tag as a marketing tool on my blog (http://kidpub.blogspot.com).

The description meta tag is configured to emit about twenty words that act as the call to action. It is typically displayed beneath the link on Google. We want the title to tell the reader in a few words what the site is, and then give them a reason to come to the site with the description.

These are the only changes made to the site during the measurement period from August to December of 2008. Granted there are external factors (word of mouth, appearance in a blog etc) that bring traffic to a site, but I'm convinced that the optimization effort is primarily what led to a doubling of traffic. We are projecting another doubling in the next three months to 20,000 unique visitors per month based on the current rate of growth.

I am currently blogging quite a bit on SEO, mainly because it's on my mind and I want to learn by sharing what I've observed. The blog is at http://kidpub.blogspot.com, and KidPub is at www.kidpub.com. I invite everyone to visit both and have a discussion on better ways to optimize Drupal.

Results of SEO Tweaking: 2008



I started seriously looking at SEO optimization on KidPub around July and August of 2008. After implementing modifications to the Drupal framework of the site (NodeWords, Meta Tags, etc), and adjusting the messaging with desription and title tags, I started seeing the results shown in the Compete.com here. Note that the optimization resulted in a doubling of unique visitors over a period of about four months.

Wednesday, January 7, 2009

Meta Description and SEO: Your Big Marketing Voice

I seem to be stuck on marketing with Google search results. With good reason, though...if your web site is like most, the majority of your traffic comes from Google.

In a previous post I discusses the importance of the title tag and how Google uses it in its display of search results. Heat maps show that most of the time spent looking at the search results page is on the upper-left part of the page, which tends to be the top three to five results.

The title tag, if you have one, is used for the link to your web site. Directly below is about 25 words of text that can come from the page being indexed, or from the meta description tag, if it exists.

If the title tag becomes your billboard, the description becomes your call to action. This is your 25-word chance to give readers (who spend just a second or two looking at the result) a compelling reason to visit your web site. Use action words here, and give the reader a reason to click. For KidPub we use the words 'visit', 'play', and 'read', and 'write' as well as 'safe' and 'fun' since our target is not only children but their parents.

The description goes into the meta description= tag. Take advantage of essentially free marketing by paying attention to how Google sees your pages, especially the title and meta description tags!

Monday, January 5, 2009

How important is the page title?

KidPub is built on the Drupal platform. I hadn't given a lot of thought to page titles and how they are used in marketing, but I now think that they are incredibly important.

The text in the title tag of a page are what shows up at the top of your Google search result. Drupal out-of-the-box doesn't do a very good job of filling this in, especially if you are using the default naming convention for posts. Heat maps of the Google search results pages show that the eye gravitates to the upper left of the page, and concentrates on the top line of each result. The title of your page may be the only thing that a potential visitor sees on Google.

Let me repeat that: The title of your page may be the only thing that a potential visitor sees on Google.

What you don't want here is some random text that Drupal or Wordpress or your staff web designer has inserted into the title tag. The title should tell the Google search user, in seven or eight words, what the site is about. When I took a hard look at the way KidPub was showing up in search results, I found everything on that first line from 'I hate Twilight' to 'Advice on avoiding homework' to 'Welcome!'. Not very marketing-friendly.

I created a ten-word description of what KidPub is and used the Drupal Nodewords module to insert this phrase into all title tags. What you see now in a Google search result is:

KidPub : Publisher of Stories, Books, Poetry For Kids, By Kids

This tells the reader exactly what we do, and it includes keywords that generate significant Google search traffic. KidPub is the first result returned for a search on 'kids publisher', for example, and also for 'stories by kids'. There's no 'and' between Books and Poetry because it causes the phrase to exceed the character limit used by Google, which means the phrase ends abruptly with a ...

Take some time to look at your page the way the Google web crawler sees it. Your title tags are a significant marketing tool and shouldn't be ignored.

Sunday, January 4, 2009

Keywords and Google organic placement

I did an interesting analysis of the meta keywords and description used on kidpub.com and how they relate to marketing with Google search.

About half of the traffic on KidPub comes from Google, but until recently I really hadn't done much in the way of marketing; mainly I'd monitored search results for a set of keywords related to the site and simply tracked their position. Much of my interest was prompted by a Drupal module called Search Referers that produces reports on the search phrases that are used to bring traffic to the site. I started to notice that the keywords I'd used in my meta tags were not the ones that were being frequently searched on.

I have about ten keywords in use at the site. Since I'd used Google AdWords in the past, I had a few old ad campaigns sitting around, and I used AdWords' traffic estimator to see what kind of monthly traffic each keyword generated on Google search. For example, one of my keywords was 'poetry'. Using the keyword tool, I found that the word 'poetry' generates about 3.3M searches each month. Great!

BUT, when I checked 'poems', I saw that the word generates about 9M searches per month! Simply by changing my keyword from 'poetry' to 'poems' I can tap an extra 6M monthly searches.

And so on for the rest of my keywords. Nearly all of the keywords I'd originally implemented had synonyms with higher potential search traffic. 'Stories' generates 16M, 'Story' generates 25M. 'Kids books' 25K, 'childrens books' 1M.

Google does serve up search results using synonyms, and the keyword tool will give you some insight into what they are, but tweaking keywords in the meta tags and title tag of the site just makes a lot of sense.