Contoleon.com » What Gets Clicked, What Gets Read

What Gets Clicked, What Gets Read

Anthony — Wed, 20 Apr 2011 10:58:16 +0000

Writing about how to grow a blog’s audience is popular. Subjects like writing top ten lists, promotion through search, social media, guest posts, submitting links to social bookmarking sites and more are often covered. Repeatedly. Advice and services for getting readers, either in bulk or from a specific group, is an industry of itself. Among the posts, tweets, videos and status updates there is one consistent theme, and it is content. Content matters. So, how do you discover what content is working for you?

Content Audit

What Content Works?

This blog is not popular or successful by any real measure. I have been writing for it for over two years as a sounding board and to attract people who are interested in the same things I am. The posts are long, do not include many lists, have a few graphs and generally don’t follow most blogging best practices.

What is the result of this approach to blogging, how does this affect what returning readers read versus what gets the attention of newcomers? What is the most popular content?

Overall Top Five Blog Posts

The top five blog posts from the last 12 months by unique views include a mix of graphs, Brisbane floods, a local football team and a MMORPG PVP/AdWords comparison.

All this list reveals is what was looked at by the most individuals. It sheds no light on whether they liked it, if they read it, or if they muttered ‘wanker’ under their breath and left. It also won’t tell you if it attracted new visitors, or was only viewed by regulars.

What type of visitor?

How new visitors are attracted and what are the differences in what they read compared to an existing audience is a good question. The relationship between the content that brings people to the site, and what they actually read can tell you a lot about what works and what doesn’t.

Returning Visitor Top Five

New Visitor Top Five

The posts that attract the most new visitors are different from those viewed by returning visitors. The two lists only share one post, and of the New Visitor Top Five, only two do not relate to recent events – either the Brisbane floods or the local football team. New visitors are more responsive to topical content, while the returning visitors came back for more posts relating to a small number of themes.

What did they read?

In this case, finding out which content held their attention is more useful than what was viewed by the most people. Google Analytics’ bounce rate is often used to determine what content engages. However it is most valuable if you want to know how many people clicked another link. It won’t tell you if they read the post and then left, or if they exited on seeing the wall of text.

Google Analytics Advanced Segment

The lists below were generated using a number of Advanced Segments, selecting traffic by type and a visit duration of greater than 60 seconds. The aim is to determine if the content was engaging and matched their expectations, motivating them to read it.

Overall

Once visits shorter than 60 seconds are removed, a number of posts from the unfiltered New Visitor list no longer appear. This implies that most of the posts that attract new readers don’t always get them to stay.

Returning Visitors

Out of the top five posts that returning visitors viewed, four of them also appear on the top five list of posts viewed for longer than 60 seconds. On the data here, there is a strong link between what this group will visit, and what they will read, at least for a little while. The same is seen for new visitors as well, at least for the time frame chosen.

New Visitors

Taking it further

There is so much more that could be done from here. At best, this is a summary, an indication that there is something to look for. A number of things to follow up on, or further questions to ask are:

Use a larger number of pages
Examine behaviour for different traffic sources
- Do new visitors follow more links?
- Do returning visitors leave once they are up-to-date?
Depth of visit by group?
What is the median time on page for each group?

Google Analytics can provide a lot of information. While there are some limitations to what it can tell you, for broad analysis of trends it is more than adequate. The most important thing is to know what the right question is.

Finding the Negatives in AdWords

Anthony — Sun, 23 Jan 2011 10:27:34 +0000

Unless you only use Exact match, you will get traffic you did not want through Google AdWords. Using either Broad or Phrase match types means giving up some control over what your ads appear for in exchange for assistance from what is arguably the largest database documenting human behaviour on the planet. There is a tool to help you manage this: negative terms.

Google is on record saying:

“Twenty percent of the queries Google receives each day are ones we haven’t seen in at least 90 days.”

Broad and phrase match makes it easy for an advertiser to deal with this and keep the size of keyword lists under control. Both of these match types can also produce irrelevant traffic. The terms added to the campaign may be similar or frequently associated with the keywords in normal search. This does not mean they are always relevant.

Irrelevant queries are not the only reason traffic might be unprofitable. Issues such as competition, pricing, geography or other circumstance can render some queries too expensive to be profitable, or even accessible for your campaigns.

Setting negative terms is the easiest way to reduce a campaign’s exposure to bad traffic. A negative term prevents an ad from showing for search queries containing it. In campaigns using terms relevant to multiple industries, negatives make a difference to the cost per acquisition.

Finding the Z for X and Y

Finding negative terms for a campaign is an interesting exercise. Unproductive clusters of terms can be found using any tool that allows for filtering by words and analysis of relevant metrics.

Simple hierarchical clustering dendrogram

For example: A campaign is over target CPA, and features two keywords, ‘A’ and ‘B’. Both keywords are using broad match. Historically these terms have generated sales and a consistent level of traffic. A search term report showing all search terms for keywords ‘A’ or ‘B’ reveals a number of other commonly occuring terms for which the ads display. Each of these terms above a certain threshold can then be analysed.

A quick analysis of other terms that account for a certain percentage of impressions (or other metric of choice) has revealed that traffic with the term ‘Z’ frequently exceeds the maximum allowed CPA, whilst other significant terms like ‘X’ and ‘Y’ are on target. So what now? Should it produce minimal sales, it should become a negative term. If it does produce some sales, the traffic for the term can be analysed again, to identify if there is another term that should be removed from the campaign.

As the number of significant terms in a campaign increases, so does the complexity of the analysis. In the example attached, the campaign’s targeted CPA on reporting period has been revised, and it needs to be adjusted to match the new target.

Campaign Term Nodes at higher than target CPA

The dataset used in the example contains a range of descriptive terms that relate to locations, product description, usage and alternative products. In this example, hierarchical clustering reveals a few terms that appear across multiple nodes. Each node is split until it reaches a threshold, after which the data would cease to be statistically significant.

The Intersection of Clusters

What you are looking for is the groupings that have the greatest impact on the metrics the campaign is measured on. If it is return on investment, the CPA is used for assessing the clusters of terms; if it is time on site, or even raw volume, those would be used instead. While this can be done within Excel, across multiple worksheets, the information is best dealt with as a network, and visualised as such.

And how to get it all wrong

There is one huge error that can lead to otherwise productive terms being discarded: decisions made without enough data, either as a result of a small dataset, or overfitting of terms. Search Engine Marketing traffic is very prone to volatility at low volumes, and the problem becomes more pronounced the lower the chance of a sale per click is at the ad.

Bing in Australia, a few weeks on

Anthony — Sat, 27 Jun 2009 05:57:08 +0000

Bing has been live in for a few weeks now, and it is time to look back and see what has changed. In short, not much. I have noticed a small dip followed by a spike in traffic on a mature site I work with. The dip in traffic was matched by a drop in terms for which the site was visible. I can only assume that the increase in traffic is due to both the increased attention Microsoft has drawn to their new search engine, and an increase in the terms that the site has appeared for.

Interest in Bing

On another domain I have noticed that terms related to Bing and Bing.com.au have produced a little traffic. This is only to be expected due to the increased interest in the search engine.

SERP reshuffle

One thing that I noticed that is worth watching are changes in the order of sites on the Search Engine Results Pages. For a few of the terms that I have been watching, the top three or four results have remained fairly fixed since lauch. However there has been a bit of movement amoungst the mid-page results. This was even evident on terms where the level of SEO activity can be assumed to be low, and with static sites.

Bing and the Search takeaway

It is simple; there is a lot to be learned right now in how the Search Engine Results Pages have been changing for low competition terms. For now, the changes on these kinds of SERPs are more likely the result of spider and algorithmic activity rather than general updates or targeted optimisation.

Use and abuse of statistical analysis

Anthony — Sun, 21 Dec 2008 10:39:24 +0000

A vegan recently tried to convince me of the health merits of his lifestyle over mine by quoting a study. The main point was that when comparing a certain set of markers associated with good health between a population of vegans and the general population, the vegans scored better. Obviously, this meant that being a vegan was better than eating meat.

I remain unconvinced. At first glance the figures seem very convincing, in a way, but they do not stand up to scrutiny. The biggest flaw is that the study is comparing a population with a high awareness of what they eat to a general population that range from meat eating health fanatics to sedentary fast food consumers. It is not a focused study, and all it really says is that people who follow a consistent diet without most junk food do better than a sample that doesn’t. For the study to be convincing it needs to start by comparing two like groups where one eats animal protein and one doesn’t.

Unfortunately, the use and abuse of statistics might as well be regarded as best practice. When it comes to influencing decisions or creating a story it is painfully effective, and unfortunately most people do not have the knowledge needed to call bullshit. What is frustrating is that often the data used is obviously poorly selected, gathered with little thought and misrepresented to produce the most bizarre insights.