Let’s Look At The Ethics Of Some Sponsored Posts!
First of all, I’d like to begin this post by saying that I LOVE SearchMetrics. I love what they do, I love their methodology and approach and I use the tool daily – it is without doubt in my top 3 list of tools / software which I could not live without. However, it can also be a major thorn in my the side. Although the SearchMetrics suite offers a lot, I want to just discuss one thing, the Weekly Visibility Score.
People are hardwired to see a graph and interpret it in seconds, especially when it comes to a possible performance metric. An upwards trend is great, and downwards trend is terrible – we all fall into this trap. Busy marketing directors and company owners are particularly vulnerable to this trap, and that is why there needs to be an emphasis on the MARKETER to interpret data accurately and communicate it effectively. Like all software and data out there, it’s really only useful and insightful if you understand the finer details which go into collating the data in the first place. This is the issue – very basic interpretation of SearchMetrics visibility trend data can paint an entirely misleading picture. Your visibility score could easily be going through the roof, and yet your fundamental campaign KPIs (visitor numbers for example) could be stagnant or even going backwards.
SearchMetrics Visibility Calculations
There are two keyword databases SearchMetrics use when rank checking. These are defined as longtail database and a weekly database.
The weekly database contains far less keywords than the longtail database and it is the database which powers the weekly visibility trend line. The Visibility Score and trend line we all know and love is calculated using 3 key factors:
– Keyword Rankings
– Keyword Search Volume
– Nature of Keyword (defined as “navigational” (brand searches) or “informational” (non-brand searches)
SearchMetrics track approximately 100 million keywords in Google (for your country of choice) throughout the period of a week, and apply the logic bullet pointed above to give a score.
The long-tail database contains approximately X times the number of keywords as the weekly database. Your sites visibility on these keywords is NOT factored into the weekly update trend line or visibility score. This is a very important distinction! If you do want this level of detail, just take a dump of the keywords shown under the long-tail section, and run it through your own tracking software.
• Number of keywords tracked – SearchMetrics tracks just 100 million keywords, the individual queries with the most search volume behind them. Google handles billions of search queries in the UK each month, approximately 15% of these have never been searched before. The longer-tail graph for search queries is extreme. 100 million keywords is not particularly significant in the context of the complete picture.
• Keyword Databases – As mentioned above, the longer-tail database used contains far more keywords than the weekly database, yet plays no role in influencing the weekly visibility trend line. So the key number of keywords tracked under “long-tail” (below) needs to play a significant part in the analysis.
• Paid Visibility – The Paid visibility score is very unreliable indeed, no fault of SearchMetrics – it’s simply an incredibly difficult metric to quantify given the inherent fast paced nature of paid search in comparison to organic search.
• Rankings fluctuate – a lot. This can skew the weekly trend line significantly, particularly if you rank for a generic keyword with considerable search volume behind it. In such cases, a drop of a few positions can see your visibility score plummet. Imagine you drop from 13th to 25th position for “car insurance”, the chances are that it will have little impact on traffic, but your visibility score will be heavily hit. If you have a number of generic keywords to prop up your visibility score, this can exaggerate the impact further. This is often the case with smaller sites. This can lead to what I like to call a “spikey graph” (below):
• Traffic Vs Visibility – As touched upon above, visibility does not equal traffic. Increased visibility does not equal more traffic, and decreased visibility does not equal less traffic. A flat line could easily mean a sites organic traffic is booming or dying, this is something we see a lot of day to day.
• Far better KPIs available – You should not be relying on the Searchmetrics visibility score to monitor your own campaigns search visibility performance (or even monitoring the close competitors performance). This can be done so much better using many of the ranking software packages available.
• Looking at general trends – Particularly useful with bigger sites. However, it can also paint a picture for smaller sites, if you keep in mind the points mentioned above!
• Penalty Identification – If you have a penalty or multiple overlapping penalties, the SearchMetrics data can be invaluable when diagnosing the roots of the issue.
• Keeping one eye on your wider competition – For example, a huge number of retail sites have benefited after Ebay picked up a penalty. When this is picked up by the industry, it is usually through SearchMetrics (although there are other services doing similar!).
– How many keywords? Look at how many keywords the data is based upon – the smaller the number the more volatile it will be.
– Which Keywords? – Look at the keywords which are powering the data, again – big generics can lead to volatile data.
– Long-tail Figure – You need to keep an eye on this figure, in many cases it can actually be a more significant KPI than the weekly visibility figure.
Let’s Look At The Ethics Of Some Sponsored Posts!