Is there any money left?

Having previously written about the Google Adsense Heatmap, I was keen to try out adsense blocks in more successful locations.

Google Adsense Heatmap: darker colours are ‘hotter’ areas

My first opportunity came with a project of my own to develop a custom Joomla template for a new website about chocolate, ‘I Dream in Chocolate‘.

Website Screenshot showing Adsense positioning

I planned the main Adsense location to take advantage of the areas highlighted in the Google heatmap, and created the Adsense blocks in Joomla using mod_html modules.

I didn’t go for the ‘hottest’ area, which I think is between the main title and content, as I thought this location might be over complicated to manage separately for different content items, and also less nice to look at. The current design adds a kind of white chocolate stripe to the page layout, though it does push the main content down a bit far, especially as I also added a link unit right above the top menu.

By keeping them near the interactive options and menu items, I hoped to take advantage of the visitor’s attention (and mouse) being in that area. At the same time, by using a different style, I have avoided them being actually mistaken for menu items and on site links (which goea against Google’s guidelines).

It does seem to be paying off so far: the following screenshot shows clicks for this new site have already overtaken my more well established site Stairway to Devon, which has the Adsense more out of the way. Bear in mind, many of the page views for the new site will have been caused by me working on it, so it’s actually been more effective than the figures suggest:

Screenshot showing Adsense results for different websites is gone: can’t show the actual numbers - it’s against Google’s rules unfortunately!

Inspired by these more encouraging results, I decided to add some better AdSense locations to this blog while upgrading WordPress at the weekend, following the excellent instructions in this link: ‘Upgrading WordPress‘ and this one: ‘How to add Adsense to your blog‘

Adding AdSense to a Wordpress blog involves editing the Wordpress PHP template files. I had some odd experiences, with AdSense blocks sometimes not appearing in some locations, or only appearing after several minutes (which I think is a delay on the Google setup end). It also required a lot of fiddling about with the CSS. But eventually I was able to get the following units set up:

1. Link unit at the top of the screen
2. Adsense Referral link between the page header and content
3. Google Search box in the sidebar
4. 2 Adsense content units in the sidebar: first a small one, then a tall one
5. Adsense content unit in the top ‘hot spot’, between the content title and paragraph text on the single post pages (which don’t have the sidebar)

While upgrading, I also tried setting up the options to ping WordPress’s list of practically everything, which should promote the blog a bit more widely, so I will be interested to see how these changes all work out.

(The first couple of months, warts and all)

Click to view all Stairway to Devon posts

Background: I’ve been working on Stairway to Devon, a Devon directory with a focus on promoting arts and environmental concerns alongside general clubs, blogs and business listings.

Summary: Once the website got to the point of being presentable enough, I started gradually making it more available, while continuing to work on the site, and beginning to actively promote it. This is how I did that and what happened.

July 2007
Started leaving it online for search engines to index. Created a Google XML Sitemap to help with this and opened a Google Webmaster Tools account.

End of July - Linked to it from a news item on my company site’s home page, offering free advertising to customers.

Mid August - added links to it from my customers’ websites (after asking them of course!)

25/8 Ordered business cards and postcards from

30/8 Applied for affiliate programmes via TradeDoubler and others.

31/8 Added it to my online portfolio of work

Comments with Hindsight:

With hindsight, much more testing was needed before linking to the site from anywhere. However, the budget (0!) did not allow for hiring someone else to test it. I felt that I had tested it thoroughly myself, without realising the extent to which as the developer I had developed blindspots relating to possible ways the site could be used, my own skills and assumptions, my own knowledge of the site’s functioning, and so on. Some of these issues may have been picked up before launching if I had paid or persuaded an outsider to test the site. (I tried persuading several, and that helped me fix several usability issues and bugs early on, but most people I approached, even with offers of freebies, were so busy that most of them never got round to it, and a few did test something, but with the best will in the world, the time a volunteer can offer is minimal (though still useful and appreciated - these few people’s help found the most basic problems early on!).

Without paying for testing, you are in a difficult situation: your earliest visitors are more like guinea pigs than they may realise, and they are unlikely to tell you if anything went wrong unless they are charged for it. I had to figure out many of the early problems using web statistics, and found the visitor tracking provided by my javascript statistics from Statcounter, were the only clues I had to how people used the site and why certain things were going wrong.

Regarding banners, adsense and monetising the site in general, many people advise filling a directory first and then thinking about how to monetise it (others disagree). My own opinion is that if the site is ready enough and the space is there for advertising (without conflicting with usability), you might as well get some professional looking adverts appearing in it, as this demonstrates the advertising potential to other advertisers that could purchase ad space directly (which is very unlikely before the content and visitor numbers are there). There’s also the possibility of earning something, however small: as an example, in the first few months this site made nothing from affiliate commissions, but a small amount began to appear via Google adsense. Around Christmas, we saw a small amount of TradeDoubler and Amazon Associates commissions. None of this is money you would leave in the street, however, it’s clear that the main income from a directory would come from either selling adspace directly or charging either for listings or for some aspect of the listings (featured listings being a common option). For either of these options to attract buyers, a directory needs to reach a critical mass of pages listed and readers visiting (which the pages bring in via search engines). To keep visitors returning, listings must also be of a reliable enough quality to maintain the directory’s good name.

From my own point of view as a web developer / designer, the site has paid for itself far more in providing a more technically advanced addition to my portfolio, a means to offer free advertising to my customers, and an opportunity to learn, gain experience of running a directory, and keep improving on new skills.

So to summarise the earliest stage: if a web project has any kind of budget to spend on it, earmark a large proportion of it for hiring outsiders to test it. Also, get the basics of the site’s earning potential in place - this can always be adjusted later. Install detailed web statistics, so you can get an idea of how people are using the site. After testing, the most important issue for a directory is to get enough quality listings in, while it’s important with any new site to keep a close eye on the visitor statistics in case common problems are preventing potential customers using your site.

The next installment of this blog will discuss the early steps I took to get listings in, and the first experiences with outsiders using the site.

For the last few months I’ve been told by Google to put more ads on my pages. This is the type of feedback report I get from AdSense:

May 2007 Optimization Report

Dear Publisher,

Here is your optimization report for the month of May. After an automatic review of your sites, we think you might be able to improve your monetization using the following tips:

Placing more than one ad unit on a page often generates more revenue.
Many of your sites only contain one ad unit.
How can I fix this? Dismiss this tip.

We hope these tips are helpful, and encourage you to experiment using different layouts and formats–no two sites monetize the same way!

Sincerely,
Google Adsense

The link from ‘How can I fix this?’ has some guidelines as which which types of page can do best with more adverts. It also raises the interesting point that the first group of ads to appear in the page’s source code will be higher paying than subsequent groups:

Maximize ad space with multiple ad units

Multiple ad units can help optimize your performance by leveraging our large inventory of ads. You can place up to three ad units per page (in addition to one link unit and two referral units per product). Remember that the best way to measure the effect of multiple ad units is to examine the impact on your overall earnings. Multiple ad units may prove particularly successful for:

* Pages with lots of text, requiring users to scroll down the page.
* Forum or message board pages, particularly within threads.
* Pages where only smaller ad formats (such as the 125 x125 button) will fit.

Tip for maximizing multiple ad units: make sure that the ad unit with the best placement on the page is the ad unit that appears first in your HTML code. This will help ensure that your prime ad real estate is occupied by the ads that place highest in the auction and will generate the most revenue for you.

Best of all, they then lead you to this page, Where should I place Google ads on my pages?, which includes a ‘heat map’ to show which areas of a web page are most successful for ad placements:

(Google heatmap showing the most successful areas for advertising)

In general, the most successful areas for adsense appear to be the places where you’d look for navigation menus and subject headings. This seems like it might conflict with usability, but there’s also a nice block in the centre just above the footer, so I’ve taken their advice and started using that one more.

But I couldn’t just take their sensible advice. I’d been putting together a resources section, but it wasn’t yet organized beyond a single page list. So I thought, let’s try out a monster page to see how many adverts a single page can stand. It’s got the maximum amount possible of Google Adsense: 3 content units, 1 search unit and 2 of each kind of referral unit. But not only that, it’s got affiliate banners and most of the text links are affiliate links as well. And to give it more content as well, it’s got an RSS feed. It’s a dog of a page - it takes a hundred years to load! Click here for a page with an insane amount of advertising!

I’ve since organised the resources into a much more sensible arrangement. Ironically, the total amount of adsense on them now is much more - in some cases, where the categories don’t have as much in them, I’ve even used the content units to fill in gaps until I’ve got more to put there.

I’m not that bothered about having to change things around, so it’s all an experiment. But if Google send me the same report next month I’ll be wondering why.

A review of 3 months of posting on Hedir

Hedir is a busy website based around a peer reviewed directory, with a community of developers, moderators and reviewers and a much larger group who submit websites for approval in their directory. The idea is that people who submit websites should also review others, but not many of them seem to catch on to this, although it’s a great way of finding out what works and what doesn’t work on other people’s websites. The Hedir website also offers forums, and blogs for its members.

Every page on Hedir shows a couple of units of adverts from Google adsense, and they operate a revenue sharing scheme which I described in my earlier post ‘Adventures with Google Adsense, part 2′. It seems on the surface of it like a good idea and very cleverly done. However, as I discussed, after a month of fairly intensive posting, I had not made any money from it. As I’d been away for 10 days in the middle, I thought for a proper trial I’d give it another month of posting to review pages and forums. By the end of this I had posted about 300 times in total, and I think that gives it a very good chance of working if it’s going to.

The Alexa traffic rank of the site seems good, at around 32,000, but there are currently about 180 pages of sites listed in the review queue. I thought perhaps this meant that each page was hardly ever viewed. If they could bring in some way of requiring everyone who submits a site to do a certain number of reviews, that would bring the size of the queue down and get each page viewed more frequently, but it wouldn’t necessarily get more views for each member’s adverts.

Another reason for the tail off in Hedir adsense views when I was away is of course that once another review has been added to each page (which is presumably the reason for most of the page views), those ads won’t show up for my adsense id any more, meaning that those pages have been removed from the number still available for earning me adsense commissions.

Given the way that people use Hedir review pages, I think there is not likely to be much commission from revenue sharing ads, at least certainly not on the review pages - I can assume this was a typical couple of months. Forums may be viewed more often, being publicly available and linked from the front page, but forum ads are only linked to the last poster 20% of the time. They also offer blogs with adsense revenue sharing, an opportunity which I haven’t tried yet.

However, I did discover another benefit to posting on Hedir: people will sometimes follow the links in your posting’s signature. So I now have about 300 signatures posted on various Hedir pages, which are all currently linking to 3 of my customers’ websites, and I’ve been seeing more traffic coming to the websites from them.

Some members also think it may have an effect on the linked sites’ Google page rank, as the links have been known to show up in their Google Webmaster Tools accounts, but I don’t know how you could check whether there is specifically a page rank effect.

In which I try out advertising somewhere less out of the way.

Not surprisingly, placing the adverts on a few out of the way pages that were only obscurely linked from my home page (if that!) meant that they were not loaded and viewed very often (ie there were not many ‘page impressions’, let alone ‘clicks’). Here’s a typical day (see the tumbleweed blowing through):

I thought to try it out properly I should put the ads on some more popular pages. After all, I’ve seen messages from people who only used to make $900 a month from AdSense until they followed certain methods they’d be happy to share, and now they make over $15,000!

As luck would have it, I had submitted a couple of websites to a peer reviewed directory called Hedir, a site which I know receives a constant stream of submissions. I discovered it is possible for users to add their Google AdSense publisher IDs to their Hedir user profiles. (See ‘Hedir Review: Community Rewards’).

What this means is that whenever I review a website submitted to Hedir or post a message in their forums the web page runs a Javascript program so that a certain percentage of the time the Google ads belong to my publisher ID and the rest of the time they belong to the Hedir owners or developers. This seems like a clever idea. Having seen that hundreds of websites are submitted to Hedir all the time, I was sure this idea would be a winner.

AdSense also provides a feature called ‘channels’ which allows you to track which pages are loaded and which ads are clicked. So I added my publisher ID and an AdSense channel ID to my Hedir user profile, reviewed about 50 websites and posted some entries into the forums. Then I checked in with Google AdSense every day for the next few weeks to see what was going on.

Here’s how the figures break down for February, my first month with Google AdSense:

Woo hoo! I made $0.13 ! It turns out that was a click from my Valentine’s Day poem. So I’m up there with Hallmark, making a profit from Valentine’s Day.

And here’s the break down for the first half of March, when I was away for 10 days so presumably not loading my own website as much:

Looks like the Hedir pages are not viewed as much as I expected. People seem happy to submit their sites but most can’t be bothered to review anyone else’s. Also, apparently people don’t click on ads much. But although I haven’t made any money from it yet, I like the idea and I’ve learned a lot more about web design from reviewing all these websites and seeing other people’s comments about them. And who knows, the pages will probably be around for a while so I could make some money from it yet.

In which I try out artificially intelligent pay-per-click advertising
After my initial experiments with affiliate marketing, pay per click advertising seemed like the next logical step. The basic setup is this: you sign up for Google AdSense or something similar (address at the top of the page), use their online advert design process to create snippets of Javascript code, and then paste this code unaltered into the appropriate part of your web page. The code then loads when your web page does, and Google’s server processes the other text on your web page and/or website and selects relevant adverts to display in the colours and layout you have chosen. You’re not allowed to click on your own adverts, but if someone else does you make a small amount of money. So millions of people click on the ads and you make an effortless fortune. Right? Well, so far I’ve found the system works apart from the last part.

The first thing I was interested in was how the adverts were chosen to be relevant. Would it depend more on the content of the page, or the website as a whole? I also didn’t want Google ads for my competitors appearing on my home page, although I did see the appeal of taking some of their money if I was booked up anyway. So I tested out the ad selection process by setting up my first Google ads on a mix of obscure and slightly odd pages to see what they came up with.

Check out how the ads appear on:

• my privacy policy
• my ethical policy
• my copyright / credits page
• my legal document templates

Next I tried:

• my ‘Under Construction’ page, with free photos of diggers.
• Then I discovered this was probably against their ‘no error pages’ rule, so I moved it to my Construction Page Hall of Fame

Then I thought of some types of work I was too booked up for at the moment:

• Content Management Systems
• Database Websites
• Internet Business Cards

While keeping an eye on the Google forums and looking at how Google AdSense is used on the net, I discovered several gimmicky pages along the lines of Find Your Hobbit Name that were designed to keep people reloading and presumably seeing new ads that they might click on. So I wrote this one:

• Random Valentine’s Day Poem

How romantic.
I even tried it on a ‘Lorem ipsum’ page, before realising that would be against AdSense policies by placing them on a page with ‘content primarily in an unsupported language’. For the record, they were mostly about beach resorts, for obvious reasons, and temp agencies, probably because of the ‘tempor’ word fragment.