Progressive tracking mechanisms allow you to track virtually anything on your website. Obviously, sending all the clicks and events into Google Analytics is only part of the story, you should also clearly understand what to do next with all these data. That's why I've decided to share examples of dashboards that can be updated in real time and used by marketers, developers, content-managers and even analysts (!) in Ecommerce projects.
Using the dashboards you can answer lot's of questions, we will have a closer look at these:
The described solution is determined by the following wannabes business-needs:
For those who are not familiar with Google BigQuery and Google Sheets, let me start with how to organize data collection and processing.
To collect and process data we use Google BigQuery. Unsampled Google Analytics data is streamed right away into Google BigQuery, and data from other services (CRM, Mandrill, Facebook, Bing) is imported automatically or manually. We chose Google BigQuery since you can import data from any service in any convenient format and mode. Plus you don't have to bother about indicators, cores and available server space.
If you have never worked with Google BigQuery, it's a cloud service that allows you to process vast amounts of data in just a few seconds, supports SQL-like syntax, and the payment is relative to the volume of data stored and processed. When registering you get 300 USD credit for 60 days and 1 TB of data processing free every month.
To get Google BigQuery data in Google Sheets, we run an SQL-query in the add-on and save it to use in future.
Yes, to pull data from Google BigQuery, you should use SQL. But there are no limitations for a report structure, number of metrics and your data is always unsampled. Anyway the queries are created only once. Google Sheets data is updated automatically or with the following query.
The table below represents comparison of marketing campaigns key indicators: number of sessions, revenue and ROAS weekly.
Please note that ROAS, unlike number of sessions or revenue, should be computable. Otherwise when changing segmentation, it will be calculated incorrectly.
To monitor the website efficiency we define key microconvertions: add to cart, checkout, transaction. We segment each of them by page type and browser.
To easily support and update reports we follow these rules:
If you need to combine a few metrics for a pivot table, you can do it the following way:
For instance, the conversion rate decreased. As a first step we check if it is related to the entire website or to particular pages. As we can see, conversion rate fell the most on product pages in Firefox browser. Developers missed this error while jQuery updating. Due to the dashboard, it was promptly identified and fixed.
Please note, that unless there is enough data for comparison, it is automatically filtered and does not distort valid data.
Besides the conversion indicators it's beneficial to monitor key technical indicators, for instance:
These metrics are also recommended for segmentation by page type and browser to spot the reason asap.
Note, that these indicators change "from the user's point of view" and this view is more representative for this task than server side monitoring (zabbix, munin).
So, how about the crocodiles? They are on guard duty, protecting data in Google datacenters.