A lot of marketers believe that setting up Google Analytics is as simple as installing the provided code on their website, in the proper place. But over the last few years the level of technical sophistication that it takes to implement Google Analytics correctly has skyrocketed and the majority of implementations are wrong. The worst part? Budgeting decisions are based off the potentially incorrect data.
Here are the five steps to creating a successful Google Analytics implementation.
- Create a Measurement Plan
The key to success when working with data lies within the foundation of your digital marketing strategy, which is your measurement plan. These measurable metrics are important in determining what portion of your digital marketing strategy is working and helps inform optimization decisions for future success.
Once you have a well-defined measurement plan, you can begin to map out effective data reporting through Google Analytics. When mapping out reporting needs, it is important to categorize the data points into two groups, dependent variables and independent variables. Dependent variables are key metrics a brand wants to influence (transactions, lead submissions, email signups, etc.). Independent variables are metrics and dimensions that have the greatest influence on key performance indicators (products, price, marketing channels, etc.).
- Review Google Analytics Account Structure
First, there is a difference between Google Tag Manager (GTM) and Google Analytics (GA). There is a frequent misconception that GTM and GA are the same, or that GTM is just the latest version of Google Analytics. They are two separate tools. GTM allows for simplistic management of all third-party JavaScript tags to be easily added to a brand’s site. GTM uses only the pieces of the JavaScript code that needs to fire to determine the action you are trying to capture on that page. Google Analytics on the other hand, is more focused on collecting, processing, and reporting insights about your website.
The goal of the review process is to identify the way GA tags are implemented on your site. This will set the stage for how flexible or inflexible tracking across every subdomain will be within your site properties. If GA codes have been hard coded to the site or added via a Google Tag Manager (GTM) it will help us understand the feasibility of data transfer across all properties.
You May Also Enjoy:
- Behavorial Analytics Habits to Keep Leads Flowing
- Marketing in a Gray Area: The Ethics of Digital Targeting
- Design an Implementation Plan
Based on the KPIs outlined in the measurement plan, do some planning to understand what this means in terms of site structure. For instance, if your KPI is a call to action on the site, it’s important to map out where this will live on the site, what the goal of this KPI is and what type of data you are hoping to get from this metric. This will begin to inform how this KPI, along with others, needs to be set up from a tagging perspective and will inform how properties are set up within GA.
- Set Up Google Analytics Properties, Views and Filters
Understanding the right way to set up properties in GA to achieve business goals and KPIs is not as easy as adding a GTM to your site, app or web property; you need to understand the structure of the account. Similar to a measurement plan, setting up properties in GA is never a one-size-fits-all solution. Every brand has a unique set of websites, mobile apps and web properties that all connect for one holistic user experience with measurable metrics.
The most prevalent mistake brands make is not looking at properties as a long-term approach to performance measurement and therefore not asking the right questions around what problems a property is going to solve. A property in Google Analytics, at its simplest form, is a website or mobile app that is associated with a tracking ID. The tracking ID is unique to that property or resource, which is what GA uses to aggregate performance data throughout the property.
The trick with properties is multiple websites or mobile apps can use the same tracking ID, because when you have multiple properties (websites, mobile apps and subdomains), you want to be able to see what a user is engaging with at entry point to conversion and everything in between.
Next is setting up views and filters. The important thing here is that the data is processed only once. If a view or filter has a problem, any data collected when the problem was live will permanently alter the data. Filters are used to modify or restrict data within a view. An example of a common filter is IP exclusions. It’s important to exclude agency or client IP addresses in data, since most of the site visits are for development, testing or auditing purposes.
- Establish Attribution Modeling Touchpoints
The last piece is ensuring data is collected in a way that supports the attribution models. Attribution modeling is a set of rules that assigns credit for sales or conversions throughout the path to conversion. While this may seem like a simple concept, if the data is not collected in the correct way, the model will not correctly assign sessions to the right channel or tactic.
The biggest challenge in setting up attribution modeling falls within how a property is set up to allow for cross-domain tracking to understand first touch vs. last touch conversion paths. It is also important to ensure all digital properties (email, social and search campaigns) are tagged with the proper GTM code for GA to assign credit to a specific channel.
With so many companies using Google Analytics tools, and every company using them for different metrics, there is a one-size-fits-all solution. It takes planning to identify the right approach, and this planning can take months or even years.
Jenny Bristow is the CEO and co-owner of Creative Anvil.
Network
