Web Analytics Planning Model for Google Analytics
Over 6 years of working with companies to implement and use Google Analytics I’ve consistently seen the greatest threat to a successful implementation is a lack of robust planning. This is the first post in what will become a series on how to build a successful strategy for adopting Google Analytics and what you need to consider when planning effective use of this great web analytics tool.
My simple model starts with business goals, moves to understanding good questions, then to what data is needed to answer those questions, which informs what implementation requirements are. Once implementation requirements are identified data can start flowing, after which point configuration and delivery of reports is required, finally delivering the right data to the right people at the right time that will answer key questions and help you to meet your goals.
For Lack of a Better Name…
I have yet to come up with something more exciting like Stéphane Hammel’s “Web Analytics Maturity Model” so I’m just calling my model a “web analytics planning process” for the time being. In this initial post I’ll explore what the key components of my planning process are and what they include. There is also an implication here for team structure, which I’ll address in a forthcoming post.
Explaining the Model
Using this model with my clients has drastically helped to clear the way for a better implementation. Why? Because the typical Google Analytics approach starts with throwing the GA tags into a site because it’s easy and free. Then, people start looking at reports and looking for the data they think they need to answer questions they have. The problem is, often that data isn’t there, and the questions they have aren’t the right questions to be asking anyway.
Step 1: Business Goals
The first step is to thoroughly understand the business goals of all stakeholders in the website and online marketing efforts. This is a large effort, but a critical one. Whether you’re a for-profit or non-profit, public or private, large or small, you have goals and a reason for existence.
By understanding the goals, i.e. what people are trying to accomplish and what you’re held accountable fore, the rest of the process and thus the results become oriented around those goals. What a novel idea eh? Business-oriented analytics – that means web analytics that can actually deliver business value, because by design the efforts are based on supporting the goals of the organization.
In addition to identifying goals, also define known challenges that stand in the way of achieving those goals. Along the way knowing what these challenges are will help you to know what to watch out for or areas where particular attention must be paid.
Bottom-line, this part of the process is all about the business, not the web. If anyone says their goals are to “know pageviews for the website” that’s not what we’re talking about here. Goals need to be mission-impacting, big picture, reason-for-existence kinds of things. The metrics will come later.
Step 2: Supporting Questions
Once your goals are identified, you can begin to define the relevant questions that need to be asked in relation to each goal. If a Goal is “increase online sales 25% without spending more money on advertising” then a relevant question could be “how to we get more performance out of our existing ad spend”. See? We’re still not talking specific metrics… those will come later!
This part of the process is largely subjective. It is about brainstorming, experience in the field, knowledge of the business and industry, website, organization, etc… Asking good questions is an art and something that everyone should contribute to. No question is off the table, but ones that are metric-specific should be set to the side for the next step.
Step 3: Data Requirements for Answering Questions.
Once you know what questions you should be asking you can determine what data you need. This is where you can finally start to throw down those requests for numbers. ”Give me my pageviews!” you’ll hear, or “what’s my conversion rate”, etc… This is the step where each goal and its supporting questions can have metrics and KPI’s defined that will relate to it.
Once you know what you data requirements actually are you can move into step 4, implementing, which is where most companies jump right into. However, by following this process you’re far ahead of the competition because you have a well-thought-out plan for what you will implement, what you need it to deliver, why you want it, and how it will help your business.
Step 4: Implementation
This is where the rubber meets the road. Conversations go from the boardroom to the developer’s desk. With your data requirements in hand you can begin to craft the technical implementation requirement that will support those data needs and then turn the developers loose on implementing those requirements. The result: a robust Google Analytics that is going to be actionable, business-relevant, and valuable. Not to mention, since you’ve involved most levels of the organization there is much greater stakeholder buy-in and support for what you’re doing.
Step 5: Report Configuration & Access
Once implementation has been completed you can finally get into the fun stuff – getting data back out of the tool. This is all about knowing Google Analytics well and working with stakeholders to setup the reports they need, either in-tool or using external applications to support your needs (my favorite external apps at this point are NextAnalytics for Excel and GeckoBoards for simple, quick dashboards).
In this step you’ll likely need to create multiple profiles, custom reports, dashboards, advanced segments, and more. The end-point of this step is having reports that are relevant to each stakeholder and answer their questions that were defined in Step 2.
Step 6: Analysis & Answers
Finally, the result of your efforts: delivering answers to those key questions that will help decision makers navigate their way to achieving goals and overcoming challenges. This is where the money comes from, plain and simple.
From Failure to Success
At trade shows and events I constantly hear from people who have usually had varying degrees of failure with Google Analytics. In almost all cases they’ve skipped steps 1 – 3 and jumped right to step 4, implementing stuff. Then they gloss over step 5 by using only the most basic of built in reports and expect to have amazing outcomes at step 6 and help meet their goal, which they never sat down to define in the first place.
Success comes from making web analytics business-relevant, and you can’t do that if you jump to throwing plain-vanilla free Google Analytics tags into your website. Success comes with a measure of discipline and best practice. My model is one way I’ve found to be greatly successful in getting there and I’m sure there are others out there as well. Bottom-line, this is essential not optional.
Next post in this series – understanding team requirements and skillsets that you’ll need for success with Google Analytics.
Thanks for reading!