Web Analytics | Lars Birkholm Petersen

Category: Web Analytics

Get better Insights with Sitecore Engagement Intelligence

November 19th, 2013 — 3:01pm

A few weeks ago Sitecore released a very exciting addition to our existing Engagement Analytics applications. Engagement Intelligence are a great choice for a Digital Analyst, as it offers great flexibility for accessing data, creating own custom Dashboards (reports) and pretty much lets you segment on everything.

So ever wondered which Persona, visitors from US that have searched for Ski and booked a flight matches most?
Now you can get that insight with Engagement Intelligence.

For this post, I have used the Engagement Intelligence Datamart (OLAP Cube) and Targit for Engagement Intelligence (BI Tool).

Sitecore have already many Dashboards available that can be used with Targit for Engagement Intelligence, so when I open Targit for Engagement Intelligence – it looks like this:
Sitecore Engagement Intelligence Overview

Let me highlight some of my favorite capabilities with Targit for Engagement Intelligence.

Access to data
You get access to Dimensions and Metrics from the Sitecore Analytics Database, which you can use to create your own rich Dashboards AND it’s easy, insert a new table, add the Dimensions and Metrics needed and you get the overview.
Create Custom Dashboards in Engagement Intelligence
(Click to enlarge)

With Sitecore Digital Marketing System – you can Own the experience and be relevant for different visitors with different needs. One powerful capability is to enable profiles (like Personas, Product interest, Stage in Customer Life Cycle etc.), to track implicit behavior on the site and match visitors to different groups that can be used for personalization.

With Engagement Intelligence you also get access to Profiles Dimension and can use that for Insights and Segmentation.

Example, which Keywords are most valuable for Parent Visitor Type?
Keywords by Parent Visitor Profile

And which are most valuable for Romantic Visitor Type?
Keywords by Romantic Profile in Engagement Intelligence

Useful insights that could be used to segment visitor by used keyword or increase spend in keywords to attract right Visitors.

Drill Down
Once you are using some of the existing Dashboards or have created your own, you can easily drill down to find insights.

Example, here I have used 2 min. to create a dashboard, with a lot of insights – I should mention I have focused from the data point of view rather than the UX (which of course always should be considered).

Just looking at this – without the drill down, it would be data puke.
Drill down in Engagement Intelligence
(Click to enlarge)

However if I click on Booking conversions, I see the other Metrics change accordingly, which gives me insights to which are my most valuable assets leading to Booking conversion, which pages are most important, which keywords was used by visitors who booked, which traffic sources was most valuable and which profiles.
Drill down in Engagement Intelligence step 2
(Click to enlarge)

Let me also drill down on Visitors who have searched for Ski and are located in US.
Drill down in Engagement Intelligence step 3
(Click to enlarge)

Now I can answer the question…
The Persona, visitors from US that have searched for Ski and booked a flight matches most are Parent Persona. Useful insights that could be used for launching new marketing initiatives towards that group.

We all know, that if we want the truth we have to ask Mom…
In Engagement Intelligence MOM is Marketing Optimization Matrix, very simple and yet powerful – on the X axis we have Visits and on the Y axis we have Value. We have divided this by to lines that represent Average Value and Average Visits, which leaves us with a Matrix.

In the Example below I have Chosen Online Campaigns for the last week
Marketing Optimization Matrix in Engagement Intelligence
(Click to enlarge)

The circles represent different Campaigns, when I hover these I can see specifics. Just by looking at my Campaigns like this, I would properly be very unhappy with the yellow circle – a Campaign where we are getting a lot of visits, but no valuable conversions – a great insight that could be used to optimize the site/Call to Actions, for visitors arriving from that Campaign.

The Dark Blue Campaigns is great, Many visits and Value – The Light Blue, is a great converter – we should properly increase traffic acquisition for that.

Merge Data
Finally we are able to combine data from different sources, so I could take my Dimension and Metrics from the Sitecore Analytics and merge that with other data sources, e.g. CRM, ERP, Ad Serving Stats and create unified Dashboards.

Example take Marketing Campaign Costs from PPC and Revenue from ERP and combine that with Sitecore Analytics about Campaign Performance.
Merge Datasources

Hope you will get a chance to “drill” in to Sitecore Engagement Intelligence, there are many insights to be found by diving into data and have this flexibility at hand.

Comment » | Digital Marketing System, Web Analytics

Oh no – can’t see the keywords from Google!

February 1st, 2012 — 4:58am

…. And still trying to be relevant for the visitor!

Back in mid-October 2011 Google made some changes, that will impact the visibility into which keywords, visitors arriving to your website from Google have used. However it will only affect visitors who are logged in using their Google Account to Google.com.

The reason for the change is “As search becomes an increasingly customized experience, we recognize the growing importance of protecting the personalized search results we deliver”, quote Evelyn Kao, Product Manager.

According to Google they expect it to be a single digit percentage of visitors, where you won’t see the keywords. However as the service, is being rolled out to various country sites and with broader adoption of Google+, that percentage could be even higher.

Reported by different bloggers within the SEO community, the numbers are constantly increasing – reaching double digits for some sites in November and are getting higher share each month.

So how can it affect you and the Customer Engaging websites, which you are trying to create and maintain?

First – a really low hanging fruit in Customer Engagement is to use the keywords for personalization, the keywords reveal a lot about user intent, like why is the visitor here, which products/services is interesting, are the visitor just searching or ready to engage?

Not having the visibility into the keywords (the intent), will make it harder for personalization based on keywords, when all you get is a (not provided) parameter.

But how could you, as an Interactive Marketer, still try to deliver an Engaging experience for the first time visitor arriving on an entry page, from Google with the keyword (not provided) parameter. The Visitor then clicks on the Home page, what would be an engaging and relevant message here?

The picture below shows the behavior of the visitor, according to what is stored in your site analytics.
How can you optimize the engagement for the visitor, coming with (not provided) keyword

So let’s evaluate which information you have access to:
- Geo information about the visitor (Country, City etc.)
- Which entry page the visitor landed on
- Referrer Search Engine is Google

This will allow you to create targeted messages based on the entry page, if it’s a specific service – the message could be – “Learn more about why these customers use the specific service”, linking to case studies around that specific service. It could also be combined with the Country or City, which the visitor is located in, so the message becomes “Learn more about why these customers in x country use the specific service”.

However the more dimensions you bring in play, the more messaging you need to create, analyze, optimize and maintain.

If you take a look at a specific entry page, just like a page about a specific service – is it clear which intent the visitor could have by looking at this page?

The clearer the more usage you could have of the above….
Unclear leads to other ways around to be relevant for the visitor.

If the entry page drives a lot of traffic to your site and most of that traffic is from Search, then chances is that the Long Tail of keywords, maybe similar to this:
The long tail of Search Keywords

In the Long tail above, 20.546 visits are coming from 29 different keyword(s) and some invisible (not provided) keywords from Google, which at position 4 are 7.8% of all search traffic leading to this page.

If we look at the top3 keyword(s) terms, they are driving 76.03% of all organic Search traffic to this page.

If your analytics is integrated with your Web Content Management System and you have access to the data in real time, just like is the case with Sitecore Customer Engagement Platform, then you can access page level analytics and get info about the top keywords for a given page, just like the top3 keyword(s) in the Long Tail example above and use this data, to be more engaging on the Home Page.

Here are two examples of what you could show on the Home Page, after the visitor clicks from the Entry Page.

Example 1: The Top1 Keyword(s) drives almost all Search traffic to the Entry Page
Then the engaging content – that should stimulate the visitor in taking the next desired action, should be based on that Top 1 keyword term, chances are high that visitor have used this Search Term to find your site.

Example 2: You get much traffic for different Keyword(s) terms
How about showing a Tag Cloud based on the Top3, Top5, Top10 used keyword(s) terms, which are driving traffic to the Entry Page, could be a simple spot that asks “Do you want more info about…. Keyword(s) Tag Cloud”. Then you are trying to be relevant and chances are (depending on how many Keyword(s) you use in the Tag Cloud), that you are being relevant. If the Visitor triggers on a specific keyword in the Tag Cloud, then you have more info in the intent and you can use this for further optimization of the visitor experiences on your site.

Some considerations:
- Not Every Entry Page leads to personalized spots on the Home Page, identify those which are important in driving traffic and where keywords often reveal some intent.
- Consider Filter “Bad” Keyword(s) sentences out, before showing them in the Tag Cloud. Worst case could be, you are getting a lot of inbound traffic on “brand product x repair”, properly not an engaging sentence to showcase on the Home Page.

Having access to data and the possibility to seamlessly integrate these data into optimizing the visitor experience in real time, is getting more and more important.

Web Analytics are not just about having access to aggregate data anymore, it’s should be a part of the real time optimization framework, which can be used to control and optimize the visitor experience in real time. The Analytics of the future is heading towards being Responsive Analytics.

Please use the comments below, if you have any comments, feedback or have other suggestions for optimizing the Home page, given the quest above.

Comment » | Customer Engagement Platform, Digital Marketing, Personalization, Sitecore, Web Analytics

Understand customer purchase behavior from Search

October 20th, 2010 — 9:08pm

Wow it’s really been a while since my last blog post, so what have I been up to all this time?

Well first and most important I became a father to a lovely baby girl, what a thrill and many new things to learn (including changing diapers)..

I also enrolled for the Web Analytics Master Certification, by Market Motive – teached by John Marshall and Avinash Kaushik – so it also takes a little time, but it’s definitely worth the time. Really inspiring lections!

It’s really cool to dig more into the wonderful world of Web Analytics – and all the fantastic insight it gives you. Combine that with the wonderful world of opportunities Sitecore Online Marketing Suite gives you, and you have some powerful opportunity, to create great results for your organization.

If you want to understand customer purchase behavior, that comes from search, you should have a look at the analytics on your site today. Segment so you only see visitors, that comes from search.

Now look at your goals mapped with different search engines and now you have an overview of which engines are bringing you most conversions. You could also segment this report by paid and organic search, or even segment brand vs. non-brand keywords – to get an understanding of you are reaching out to new customers, who don’t know you in advance (searched for non-brand keywords).

Remember to look at the conversion ratio, perhaps paid search from Bing aren’t giving you that much conversions, but if the conversion rate is high – you properly wants to throw more money at that Bing paid search.

Now you have insights into which engines are bringing you most conversions, drill down to see which keywords from each engine gives you conversions – are there any surprises here?

Are there any keywords that are giving you lower conversions than expected or even worse – they don’t convert at all! Also take a look at which landingpages keywords are routing visitors to and the bounces on them.

It takes some time, but it’s definitely worth the time! And you should end up with a lot of insight into visitor behavior, dig more into the wonderful world of data and understand visitor behavior on keywords, you expected convert but doesn’t – do they bounce, which pages do they look at, do they use internal search (what do they search for), where do they leave your site.

If you have high bounces on landingpages, you should properly revise the content, think about which call to actions there should be on the page.

Segmenting the Keywords
In terms of keywords, try to index those into three buckets:
1. Looking
2. Confirming
3. Buying

Different Target Groups and Behaviors for Web visitors

These three phases, maps the user journey from visitor to customer, if you sell cameras on your site, the user journey could be:

1. Peter wants a new digital camera, he goes into Google and search for “Digital Camera”, which gives him the Search Engine Result Page. Peter clicks and browse several of the results, learning more on Digital Cameras and finding out which camera, could suit him best. It could be, that a perfect fit for Peter is a handy compact camera with HD video.

2. Peter search for “Digital compact Camera with HD video”, reading about different brands, customer reviews/recommendations and price. Perhaps best match for Peter is the Canon IXUS 120 IS.

3. Knowing which camera is best match, Peter search for “Canon IXUS 120 IS”, here he looks at several sites – looking at price, deliverability, Gurantees, Service etc. – And properly ends up buying the camera.

The user journey above could take from a few hours to several days. If you think about the three phases – it’s really hard to sell a camera to a visitor in the Looking phase, you could properly (with the right triggers) sell to visitors in the Confirming phase and you just need to motivate that visitor in the buying phase, to buy your product, depending on that they have the ability to do so (money).

By mapping your keywords into these three phases, you could identify the visitors once they arrive on your site, from Search – and show them relevant content according to the phase they are in. If they are in the looking phase, you could focus on getting a sign-up for a newsletter or a registration, so you can start your dialogue and be top of mind along the road to the big conversion (buying). If a visitor are in the buying phase, you would properly show customer statements about your fabulous service or have quick you can deliver, even a special offer could trigger the conversion.

Your list of keywords mapped to the different phases, could look like this:

Keyword looking, confirming and buying phase analysis

Rules for personalization
Next step is to bind the keywords, to personalization rules – here you could build three simple rules with Sitecore Online Marketing Suite:

Looking phase rule, where the condition are that keywords match any the keywords from your looking phase list. If a keyword match, then show relevant content – in this case triggers for getting a sign-up with right message “Check out our comprehensive guide for selecting the right Digital Camera, click here to register”.

Confirming phase rule, where the condition are that keywords match any the keywords from your confirming phase list. If a keyword match, then show relevant content.

Buying phase rule, where the condition are that keywords match any the keywords from your buying phase list. If a keyword match, then show relevant content.

Of course you could build more rules, that reflects some of your top priority keywords, like show relevant messages for Canon IXUS 120 IS, if the visitors has searched for “Canon IXUS 120 IS”, it should bring you more value – but just by creating three simple rules, you have distanced your site from all the other static brochureware sites out-there.

Then you could simple place the rules on your design, so they would be triggered when the user comes from search with a matching keyword and sees the page with the rule. It could be a user search for “Canon IXUS 120 IS”, enters the Canon IXUS 120 IS product page, clicks on “About us” – and here the rule will show related content regarding Canon IXUS 120 IS, like “Click here to see why Canon IXUS 120 IS is best in test”.

But don’t stop here – go further and optimize on keywords, with higher Average Order Value than other keywords, build rules that are using insights on Days to purchase or Visits to purchase.

In the example above I have used purchase as buying a product on a eCommerce site, but it could be any type of site, like a B2B lead generation site – where the purchase is happening, when the visitor submit a form like “Call me up”.

Are you using the insights from Search keywords to engage with the visitor today?

Please share your thoughts or experiences, by commenting below.


1 comment » | Behavioral Targeting, Online Marketing, Sitecore, Web Analytics

Back to top