Dashboard is one of the most important parts of web analytics reporting. It can provide a quick snapshot of your website performance. It can also be used to present your site performance to anyone in your organization.
You can create dashboards using several third party solutions , or you can use out of the box solutions from analytics solution providers like Google, Omniture or Coremetrics.
Honestly, none of these solutions can provide you web analytics dashboards for senior leadership or c-level executives. People in c-level groups do not have time to analyze data or interpret multiple graphs. They are fast decision makers, visually oriented, and they need something that can help them make effective decisions immediately.
This motivated me to take a strategic approach towards designing executive dashboard for my senior leadership team. Here is a step by step approach that I take to design an executive dashboard.
Step 1. Identifying Key Performance Indicators (KPIs)
Notice, I mentioned identifying KPIs and not creating them. It is funny that you can find a ton of blog posts on the internet on how to create KPIs. The truth is you don’t have to create kpis unless you are a brand new company with a brand new website. All you need to do is identify and align your kpis with your web metrics.
Ok Sameer, then tell me how to identify KPIs.
First, meet with different department managers in your company (lead generation, qualification, sales team). Identify the department level kpis and link them to your website metric.
Once you have a list of all the important kpis, reduce to list to only four kpis that can effectively present the performance of your website.
Here are the four KPIs that I use for my enterprise dashboard –
1. Multi channel traffic
2. Site Bounce rate
3. Conversion 1 = Traffic to lead
4. Conversion 2 = Lead to opportunity
Step 2. Prepare data set
I prefer using 12 months of data for e-dashboards and the data is segmented in quarterly buckets. This helps align your reports with the corporate financial reporting structure.
Here is a quick snapshot on the data sets used for preparing executive dashboard.
1. Multi channel traffic data – Contains traffic from paid search, organic search, referrals, direct traffic, social media traffic, banner, email and affiliate links.
2. Bounce rate – Overall site bounce rate from your analytics solution
3. Conversion 1 – Overall site conversion data from analytics solution
4. Conversion 2 – Conversion data from your internal lead / sales department.
Step 3. Statistical Significance
Google recently released “analytics intelligence” feature in their free analytics solution. The concept of analytics intelligence is based on using statistical control limits on data sets.
Upper control limit (UCL) and lower control limit (LCL) can be used to identify a positive or a negative trend in data. It allows you to take immediate action on the target channel.
To calculate control limits, I use the control chart calculator. You will need the data mean, standard deviation and the sample size to calculate the upper and lower control limits on data set using the control chart calculator.
Step 4. Graphing data sets
Bar charts are very effective for presenting the traffic segment performance, and I prefer using them for the traffic KPI.
First, calculate the thirteen week average for each of the previous three quarters (Q1, Q2 and Q3 weekly average) traffic data set and align the data on excel table.
Second, align each week traffic segment data from week one to the most recent week for the present quarter.
At the start of each new quarter, you will move the aggregated weekly data to the last quarter thirteen week average row. The first quarter on your data set will be replaced by the second one and so on.
I know it may sound confusing and here is a screen shot of how the traffic graph will look like. Notice, how at one glance you will can look at annual performance of all the channels and at the same time get granular level insight on most recent data set.
This graph indicates that traffic for all the channels except the referral channel is consistent.
Our second KPIs graph is the overall site bounce rate graph. For the bounce rate, I use the line graph and plot the statistical control limit against the actual data.
Here is a snapshot of the excel table for the graph. The data set is tabled similar to the traffic graph with the only difference being the control limits.
Here is how the output graph looks like. The best part about this trend line graph is it automatically allows anyone to evaluate the performance of KPIs against the benchmark control limits.
In this case, the website bounce rate dropped 5% from the lower control limit.
Creating graphs for the remaining two KPIs becomes a lot easier now. All you have to do is replace the bounce rate with conversion data.
Step 5. Presentation
In previous steps, we successfully reduced gigabytes of raw data to actionable visual elements.
Presenting these graphs on a power point slide is going to be challenging because if you put one graph per slide then you will need four slides. Keep in mind, senior leaders are too busy, and you do not want to waste their time with a ton of slides every week.
Don’t worry my friend..AIDA marketing formula comes to rescue here. Use AIDA formula to divide a single slide into four quadrants each representing one of the AIDA elements.
You can use circle markers to identify key trends and add text box to explain core issues/success points.
Hope you enjoyed this post. Email me if you have any questions. Please share your thoughts, insights and comments because it really helps me write more effectively.