The key limitation, if I have understood things correctly, is that you currently can’t use segmentation within the API.  Which is fine when you are getting general numbers out but not when you need to create a segmented dashboard across a dozen metrics or so.  It also means you cannot get visits to groups of pages, e.g. visits which saw a product page.  However, there are a couple of workarounds.

For the first issue, you need to return to the old system of creating a profile per data segment.  So, as I have recently done, create a profile for New Visits, another for UK Visitors, one for Paid Search and so on.  Then, with your segments already created, you can easily extract top line numbers from each profile and combine to create that segmented automated report.

The second issue can be resolved through the use of goals.  Two key points to remember are goals can be created based on Head Match or Regular Expressions for page names and that they can only be triggered once per visit.  Given this, the number of goal conversions is suddenly equivalent to the number of visits in which a group of pages was viewed e.g. set up a goal for view a product page and the number of goal conversions is the number of visits in which a product page was viewed.

The API has not yet been updated with the upgrade to 20 goals so you can currently only use the API on the first 4 goals but hopefully that update won’t be far away.  Along of course, with the ability to access segmented data and also to extract the numbers from within funnel visualisations.  In the meantime, I hope these two tips are helpful.

Most metrics, when viewed at daily level, contain a form of daily seasonality.  This is most clear in metrics such as visits, page views or sales which are absolute numbers.  There is a re-occuring pattern throughout the week with peaks and troughs on the same day/s each week.  An example of this pattern can be seen in Figure 1 below.

While this makes any chart pretty to look at, it makes it difficult to really identify trends or spikes in the data.  Is a data point high because there was a spike or because it was a Monday?  It is school holidays but should the number of visits on that Sat really be that low?  And of course, what day did we start to see traffic decline from and how much of a change is it really?

Figure 1

A common method used to remove daily seasonality is to smooth the line out using a moving average.  As it is a weekly pattern, a seven point moving average should lead to a nice smooth line.  Unfortunately, as can be seen in Figure 2, this means you get a nice smooth line, hiding most of those interesting spikes and step changes and general data trends.  You can see overall trends but you cannot pinpoint particular days when a change occurred.  It is also difficult to clearly identify a change immediately, as each day only contributes one seventh to each data point.

Figure 2

What I advise doing instead is to remove the daily seasonality from each data point, resulting in a line that is unaffected by what day of the week it is.  Using this method means that it is clear to see if the performance each day was good or bad. For example, in Figure 3, it can be seen that the relatively worst day for visits was actually the 25th Aug, even though visits for that day were higher than for other days during the reported period.  The technique for removing daily seasonality can be applied each day, meaning that you can identify and react to a change in performance immediately.

Figure 3

The difficulty then is in calculating the daily seasonality across a week.  This can be done properly using SPSS or a similar tool but I use a quick hack workaround in Excel that, while not 100% accurate, gets the job done.  The steps to calculate daily seasonality for a metric (using the examples of visits) are as follows, with the example displayed in Figure 4:

1. Extract historical daily visits data.  You will need at least 6 weeks, more if the period includes a known number of factors that could impact on traffic e.g. school holidays, public holidays, product releases, marketing campaigns, etc.
2. Reorder the data so that each column contains a single week and each row contains only data for a particular day of the week.
3. Recreate this table so but replace the visits for each day with the % that visits for that day contributed to total visits for that week.
4. Add two more columns to calculate the mean and median for each row of data.
5. Delete all weeks which contain days which don’t reflect the general pattern.  In this example, weeks 5 and 6 were deleted.  At this point, the mean and the median should be relatively similar for each day of the week.
6. The daily seasonality pattern is achieved by multiplying the daily mean by 7.

Figure 4

This daily seasonality pattern can then be used for removing daily seasonality for that metric for any day.  Simply divide the value for each day by the relevant daily seasonality in order to remove it.  I generally do this using a vlookup against the day of the week for each date.

Going back to the reason for web analytics, you can use this technique to clean data so that you can instantly identify good and bad days, whether this is historical data or just for the preceding day.  If you are using this for historical data, you can identify the interesting days to investigate further (play with by segmenting).  If you are using on an on-going basis, you can see instantly what performance was like for the previous day and if need be, investigate and react to a change accordingly.

Currently, in order to be able to do this sort of analysis, you need to extract the data into Excel.  Hopefully one day, web analytics tools will allows you to upload a daily seasonality pattern for a metric so that you can display the daily data with this seasonality removed.  And my dream is of a tool that would incorporate the ability to automatically create the pattern for any selected metric (with manual over rides for tweaking of course).

The other key use that I have found for a daily seasonality pattern is it can be used in forecasting daily traffic levels.  If you are able to forecast what the week’s traffic should be, this can easily be multiplied out using the daily seasonality pattern to forecast traffic at a daily level.

A copy of the Excel file containing all the data, charts and formulae used in the examples above can be downloaded here – Daily Seasonality File.

Nedstat released its new upgrade to Sitestat, a segmentation tool, three weeks ago now. It provides the ability to filter any of the standard and custom Sitestat reports using any of large range of variables, either individually or in detailed combinations. However the big question is how to go about integrating this new tool in your web data analysis. Top analysts have probably already found a million and one uses, for most people though I would recommend starting with small steps and extending this over time.

Save some basic segments

When I first started playing with Segmentation, there were some basic filters that I wanted to try out. These are filters that I can imagine everyone, regardless of the type of website, will want to apply to many of the reports they look at, both in their regular and ad-hoc reporting. While it only takes a few seconds to set up each of these filters, setting them up once and saving them means they are very easy to use in the future. They can then be applied to the whole report or an individual report item.

The key filters that I would recommend setting up as pre-saved segments are:

• Each of your traffic sources – direct entry, organic search, external referrers, paid search, online display, email, affiliates (note that different traffic sources will require you to use alternative filters options in order to be set up)
• Type of visitor – either new or returning
• Location of visitor – keeping this simple, just UK and non UK.
• Level of visitor usage of the website – this could relate to the number of visits per visitor or the number of pages viewed. It should be set up for light and heavy users and will require you to define these levels.
• Visitors who bounce – possibly the light users above, this should be users who only viewed one page. Although a better option might be to also combine with the filters for new to the website and/or only made one visit.

Remember that Sitestat segmentation is based on visitors and therefore some of these segments will only be indicative rather than absolute. For example, the segment for paid search will include all visits and pages viewed for these visitors during the defined time period, not just data for visits from paid search.

Comparing website performance for different segments

The first way I started using segmentation was to look at a single report item across the whole website and then for various filter options. A great example of this would be to look at the browse to buy rate for alternative traffic sources in order to understand the effectiveness of your online marketing spend. Other report items that I would find it very interesting to compare across different traffic sources include the frequency table reports (visits per visitor, page views or duration per visit) or navigation reports (funnels and click path explorer).

Create a report containing the single report item but do not create any segmentation filter to be applied to the report. Copy the report item multiple times, once for each segment that you wish to look at – this would mean pressing the copy button five times if you want to compare direct entry, external referrers, natural search, paid search and display advertising. Then by editing each of the copied report items, apply a different segment to each one.

This report now displays a single metric that can be compared across multiple segments and indexed against the performance of the website as a whole. For ease of reading, it can be transferred across to excel. I can imagine managers would appreciate a simple excel table that contains the performance of the predefined website KPIs for the previous month indexed across traffic sources.

Successful Visitors

A blog post that is currently only half written in my head will deal with the concept of successful visitors. It will deal with the idea that what we should be trending and reporting on is only successful visitors/visits instead of all visitors/visits as most people would do currently. This would eliminate visits where the visitor simply bounces and ideally all visits where the visitors was actually not really interested in your website. There should then be a clear correlation between the number of successful visitors in a week and the success of the website during that week as defined by conversions or revenue.

But that is all for a future post.

In the meantime, just putting it simply, it could be worthwhile creating a couple of more complex segments that approximate a successful visitor for your website. This could be a visitor who (for example) visited at least 3 times, viewed at least 6 pages during a visit, viewed the ‘contact us’ page and/or downloaded a PDF. This segment should be refined over time but it would be interesting to look at the percentage of website visitors who are acting like you would like them to. More on this in the future.

Save both the report and the segment

Something which tripped me up a couple of times at first is that the segment that you have applied to a report needs to be saved separately to the report in order for it to be applied to the report the next time it is opened. Also, if you are saving in the report in the shared folder (e.g. so that it can be shared with a colleague), the segment must also be saved in the shared folder for segments. Given this, if you do set up a set of standard segments as I recommend above, this should be done in the shared folder.

Using the AND/OR options in segmentation

The default option within segmentation when using multiple filters within a segment is AND. This means that all conditions must be met in order for the data for that visitor to be included in the report. In order to switch this to OR, the desired filter should be dragged within the boundary of the other filter. Try playing around and you should get the hang of this pretty quickly. However it gets a little more difficult when setting up complicated segments.

For example, a segment that looks like (a OR b) AND c AND (d OR e) is quite straight forward to set up. But a segment that is (a AND b) OR c OR (d AND e) cannot be done within the one operation. The workaround for this is to first create and save two segments, one for a AND b and the other for d AND e. A new segment can then be set up that is ‘segment 1′ OR c OR ‘segment 2′ using the filter option for include segment.

Using the any/all/first/last options in segmentation

This option is relevant for the many filters that relate to visit level. Examples include the different traffic sources, made a purchase, viewed a certain page, etc. These are all options which a visitor could do in one visit to the website but not in another. You have the option of choosing whether visitors need to meet this criteria during any visit to the website during the selected time period, during every visit or during a specific visit. The default is any visit but the option you select really depends on the business question that you are trying to answer through creating this segment.

The first and last visit options could be very relevant for attributing revenue or conversions to alternative marketing campaigns. The all visits option can be useful in viewing data only for visitors that always perform certain actions. But I imagine that most of the time you will simply leave this as any visit.

Note again that this criteria is used for selecting the visitors whose data will be included in the report. Once the visitor is selected, all of their data will be included regardless of whether any or all is used.

The logic for this may seem fairly straight forward at first. However once more than one filter has been included, it can get a little tricky. It increases in difficulty with multiple filters, combining AND and ORs, combining any visit and all visits and also through using the DID NOT option. If a detailed segment will help in answering a business question, then do go ahead and create it. I just recommend being careful, possibly reading the segment out aloud as a sense check, in order to avoid misinterpreting the data that you have produced.

So you have a new free tool, it is going to be incredibly useful in understanding and improving the performance of your website and your online marketing, now how do you use it?

Creating Segments

First of all, create a new report in Sitestat, you can use any report items that you want. At the top of the screen, you should see a new line with the term ‘All Visitors’ and an arrow on the right. Clicking on the arrow opens up the Segmentation area of Sitestat.

On the left are all of the filter menu options. There are a multitude of filter here, selecting one is a simple matter of clicking and dragging across to the main area. In the same way as in Sitestat, anything that is in blue is clickable, meaning the range of filter options is actually well over 100. Once a filter has been selected, to apply to the report, just hit the ‘Update Report’ link at the bottom of the Segmentation area.

Two or more filters can be combined using AND and OR boolean logic in order to create quite specific segments. Note as well that the yellow bar next to a filter that has been selected represents the proportion of visitors that meet the specified criteria, a quick way of understanding if your filter is for many or few visitors.

It is all about User Behaviour

A key point to note though is that Segmentation is always based on visitors. This needs to be kept in mind when looking at any data, it is for all visitors who meet the criteria set within the segmentation filter. The data shown is for all measurements and across all visits within the specified time period for these visitors, not just the visits or measurements that meet the specified criteria.

This point means that you may feel somewhat constrained in using segmentation in that it does not mean you can use all Sitestat reports for just one section of your website or for just one traffic source. Instead, Segmentation is at it’s best in giving insights into user behaviour. But for whatever you are looking at, it does give indicative numbers, it does allow you to trend performance and it will give valuable insights that will enable you to make sound business decisions.

Useful Segments

The key segments that I initially anticipate using are for new and returning users, for different traffic sources and ideally, for successful visitors. The different traffic sources (direct entry, external referrers, natural search and alternative online marketing sources) will be incredibly useful segments for understanding the behaviour of visitors from each source and understanding how valuable each source is to your website. This can guide in future budgeting decisions.

However, defining a segment for successful visitors could be the most powerful use of Segmentation. This definition of sucess might include the filter options for making a purchase, spending at least a certain amount of time on the site, visiting a certain area of the website, etc. But once you have this definition, it can be used to understand the impact of product features in driving success, understand the relative contribution of those traffic sources, etc. You can move beyond a focus on the quantity of traffic to focusing on the quality of traffic.

Final Points

Again, just remember that for these segments, you are looking at all measurements from all visits by these visitors, not just the visits which contributed to the success or which were from that particular traffic source.

A couple of last points to note are that a segments can be applied to the whole report or to just individual report items (by editing the report item and using the Override tab) and that all segments can be saved and applied to any other reports.

I expect to write a couple more posts on using this new tool. If anyone wants to ask any questions regarding the best way to use it, feel free and I will try to answer in my next post.