Tuesday, December 30, 2008

I live and breath bounce rates.

I live and breath bounce rates.

I think that a bounce rate is the single best measure of the effectiveness of a page.

This is because I organize my site as a big sales funnel where each page on the site is like a step in a classic (ie face-to-face) sales call. If you organized your sales process as just one page, it would mean a lot less.

The entry pages are like a salesmen saying, "Hello, would you like to talk a little?"
The next page is like a conversation in which the salesmen and prospect discuss the product. Information is exchanged, trust is built.

The click from this page to get to the service link means that the prospect is intrigued and wants the benefits and features that he's learned about (actually, she....most of my customers are Moms).

And the sign-up page is the page that resolves all the sales inhibitors (ie, do I trust these people? Is the price too high? What if I change my mind?). If I can resolves enough of them, I can get them to enter their credit card and click enter.

Then, the real fun begins.....

Sunday, December 28, 2008

Marketing Plans for 2009

Here's what I'm thinking about in terms of strategy...

Fine tune the SEO strategy or keep throwing pages and links at it? Are there overlooked phrases? Do we have phrases that never convert that we should rethink? The answer of course is yes. But, if we really devoted two man months to this project, could we improve our performance enough to make it worthwhile? Or, should I go with my traditional approach of just throwing my pages and links at the problem?

How to reduce my credit card processing fees? I have a whole blog on my questions about merchant accounts.

Improved email marketing. Our best asset might be our lists of people who have asked for my info and our lists of happy satisfied customers. We've finally built the tools to explore and market to these lists. I'm expecting this to be the single largest engine of growth for 2009. To really take advantage of it, I'll need more products and services to market.

Mastery of advertising revenue. I'm planning to get more effective at making money from advertising. This year, we experimented with CPM advertising but not with much success. We got some nice $2-$6 rates but only on about 2% of our inventory. I'm hoping next year (thru maybe Tribal Fusion) to get $2 on some large amounts (50%) of our inventory of above-the-fold wide skyscraper ads.

Product diversification. I'm thinking that while SpellingCity is a good start, the reality is that 80% plus of the costs are programming and less than 20% is content. So, for just a 40% increase, I could probably get the site recreated for French and Spanish. hmmmmmmmmm. Plus there's the long-awaited VocabularyCity.com

Sunday, December 14, 2008

Local Search, SEO, Joomla, Revenue

The topic of Local Search, SEO, Joomla, Revenue is pretty confusing. It's a jumble. Welcome to my world. I try to think thru all these problems at once. In any case, here's a plan with some questions. What do you think?


I have a site - spellingcity.com - which is a link magnet. Schools and parents love it. They link to it. They bookmark it. They tell their friends about this great way to prep for spelling tests. I'm lucky to say it's mine. (It's also a money pit but I won't whine on about this since in retrospect, most of these software money pits are essentially due to lack of clear thinking. Which in retrospect, appears really stupid).

SpellingCity has hidden, in it's fertile joomla-php-html body, a home page for each public and private school in the country. Wouldn't it be cool if these pages came up high in the search engines?

School Page Positioning in the Search Engines. There are a few steps: Setting the pages up right, telling Google about them with a Google Site Map, and getting some links.

The Google Sitemap is a bit of a mess of a project. Joomla, you see, has a set of automatic site map generators. But, with our version and implementation of Joomla, none of these seem to work the way that we want them to. So we will probably write our own.

Setting up the Pages Right - We have (I should check if this is true) set up the pages so that the:

H1 Tag - School
H2 Tag - City
H3 tags - zip code, street address, city, state, school name

Content: All of the above plus (still to be optimized).

Somerset Elementary School has this page for it's teachers and students and parents. Any parent or teacher who would like to, can create an account on SpellingCity.com, and save their spelling and vocabulary lists for the use of the students. Many schools to help their students with their homework and studying, link directly to this page.

SpellingCity.com - SpellingCity helps build spelling and vocabulary and reading skills. Anyone can enter their spelling word list and take the tests, use the TeachMe funtion, or play the educational word games. Free! To save lists, you must register and become a member which costs between $19.95 and $39.95 per year.

There are plenty of questions to resolve about page titles, URLs, metatags, metadescriptions, and alt tags. We'll work them.

Questions:

1. If I want to appear high in the search engine rankings for the school, what is the best use of the Heading tags? Page title? URL? The problem is that since we're using joomla, we have uneven control over these elements. But it would help if we had any sense of what people have already learned about how Google is treating these issues.

2. Over time, I will put up some modest advertisements on these pages. Lets say for instance that we get Huntington Learning Centers as a sponsor. Ideally, we would have some way of providing direct links to the local centers. So, on each page for instance, there could be a generic Huntington advertisment followed by a listing or link to the sites that are closest to the school. I guess this could be done by Zip code. The same concept would work with book stores (BORDERS - want to sponsor SpellingCity?), Kumon, Sylvan Learning Centers, or, if I can't do better, McDonalds health centers or the 7-11 value stores.

One possibility for making a few bucks might be to try and put up google or yahoo local advertising. Questions:
- Does this make more generally than other Google ads?
- If I want local advertising, is this automatic or do I have to do something different?
- I have never really understood what google keys it's adsense from. I doubt many people do. But is it the same mysterious complex array of info including page title, anchor text of incoming links, H tags, and content analysis? I would guess that it's simpler. I would guess that local advertising starts with:
A. The location of the viewer
B. Title tags, Heading tags, and content

Any special insight into these challenges?

Friday, December 12, 2008

Tracking and analytics

I thought the state-of-the-art was quantcast. I was just talking to an ad network and they were pushing Compete.com. People still ask about Alexa & Google pagerank. Google analytics feels to me like a standard. Here's a quick list:

Quantcast - gets data from ISP. You can improve data by putting it on your site.
Alexa - gets data from browsers. Easily gamed.
Compete.com - No idea where they get their data.
Google page rank - The green bar is a very simplistic count of how many links in you had some number of months ago.
Google Analytics - Increasingly, this is being generally adopted so they become a uniform way of counting and comparing websites.

Other market leaders?

Tuesday, December 09, 2008

Statistics - Who Cares? You should!

I've been involved in online marketing and learning SEO for five years now as a practicitioner and I've observed a number of constantly repeated mistakes by the industry. Near the top of this list is their total ignorance regarding statistics.

I often look at great analytical methodologies which are impressive in their approach but which never deal with the problem that they often make major strategic decisions without considering whether their sample size validates a decision or not.

I've yet to see an article or discussion on this topic. I've seen people give talks before thousands of other supposedly industry gurus in which they describe methodology in great detail without mentioning sample size and statistics.

At times, I've asked and probed. The industry gurus look at me puzzled and wonder what I'm going on about.

Am I the only one who took statistics courses in college and business school?

Wednesday, December 03, 2008

Sick of domain parking?

Are you sick of domain parking yet? I am.

I don't like it because it's trying to make money by being clever without adding value. I don't like stumbling across parked domains. I don't like how much time I've wasted trying to figure out this stupid area. I don't like that I haven't made any money domain parking.

A month just ended so I got this email from godaddy. Is it worth doing what they suggest?

We want you to know exactly how your CashParking plan is doing, so each month we email you a monthly account update. Here's where your account stands for November 2008:


==============================
==================
REVENUE:
November 2008 Revenue Earned*: $34.31
RPM**: $38.86
Approximate Payout Date: January 15, 2009

TOTAL DOMAIN SUMMARY:
Total Domain Count: 74
Active Domain Count: 73
Removed Domain Count: 1

DOMAIN ACTIVITY:
Domains Added: 0
Domains Removed: 0

Account Status: Active
Domains within 30 days of expiration: 2
Active domains not pointing to CashParking Name Servers: 1
================================================


++++++++++++++++++++++++++++++++++++++++++++++++
A parked page that looks like a Web site increases the likelihood of more clicks --
and that means more CA$H! So add a one-of-a-kind page header that's sure to grab attention!

Monday, December 01, 2008

SignUp forms that slide in

I would like to try to see how they work for us. I don't know if they are javascript or what. Here's some examples:
http://www.searchenginecollege.com/index.shtml
homeschool.com's internal pages (i think after they slide in once, they set a cookie so you don't see it again)
http://www.surfnetkids.com/

note, some of them are not up all the time.
Many are programmed to only slide in the first time you visit the site (in 30 days).