Web Analytics: Making website success measurable and manageable
In a business sense, we have moved way beyond thinking that the internet is an intangible, stab-in-the-dark marketing tool. With the use of various data collected from the website and its visitors, the information can be gathered in ways that are suitable for your particular business, dealing with your particular market.
Web analytics is an extremely rich source of information that helps your business understand how well your web investment is doing, and how you might be able to do better. It’s a learning experience. And over time, you can dramatically improve your results, and you can prove it.
Improving your business using web analytics entails a continual improvement process. We will go through the steps of this web analytics process to better understand what is involved:
Web Analytics: Continual Improvement Process
Define Strategy
This comes from your business-wide goals and objectives on how the website will support the business. In other words, the reasons you have a website:
- Increase orders
- Improve retention
- Cost savings
- Generate leads
- Manage public relations
- Improve customer service
To understand what to measure, the business must first know what it wants to achieve, not the other way around.
Define Objectives
The objectives are how you plan to achieve that strategy. This requires a deep understanding of your business and your customers:
- Increase e-mail clickthroughs (links clicked in emails)
- Increase repeat purchases
- Increase new web accounts
- Increase use of an online tool. eg. an online directory, job request form
- Reduce cost of new customers
Measure
Measurement allows you to get the information that will tell you if your work is succeeding, failing, or perhaps worse, doesn’t make a difference. It allows you to:
- Adhere to goals
- Monitor performance
- Compare actual outcomes with expected outcomes
What you measure comes from the defined objectives. Measurement tells you what happened, but not why.
Data is gathered from:
- Server log files
- Browser activity feedback (via built-in click tags on web site pages)
- Search log files
- Customer data (feedback, relationship information, surveys)
- Usability testing and focus groups
This data is then used to create Key Performance Indicators (KPIs), such as:
- orders per site visit,
- cost per conversion, and
- order conversion rate per campaign
just to name a few. The particular KPIs used are agreed upon within your organisation in order to achieve the desired business goals and results. 3-8 KPIs would be included in a report for a specific recipient, and there may be a number of recipients within one organisation (eg CEO, marketing manager, customer service manager).
Analyse
Analysis looks at differences over various measures of time. Analysis mostly reveals the match between your product and your audience. You can infer customer behaviours and, if correct, predict outcomes.
Analysis involves several tasks:
- Data selection – select the correct data and KPIs.
- Data review – look over the figures.
- Causal hypothesis – what happened and why.
- Lessons learned – draw conclusions.
- Opportunity assessment – what options to take advantage of from the information.
- Recommendations – Specific actions to take.
Report
Reporting is taking the analysis deliverables and writing up the results. All too often people fail to take action on web data. Written recommendations, though, tend to become to-do lists.
For reporting to be effective, it must be provided meaningfully for the recipient. For example:
- Most CEO’s don’t need the number of daily visitors every day. Weekly or monthly is often enough.
- However, the marketer in charge of a campaign may need those conversion numbers daily (or even hourly), and monthly is just not useful.
Action
Taking action on what you have learned is often harder than it should be. The reports, and the habit of actioning them and continuing to generate them, need to be integrated into your business to achieve any benefit out of them.
Distribute your report to the decision makers. Then meet to focus on and discuss the findings, deciding what changes need to be made to the website and/or campaigns.
Create an action plan so that required tasks are allocated to appropriate people/teams, and deadlines set.
Then do it!
Measure Again
You need to repeat the process to see if the changes worked. If the changes did not improve your results, back it out. You learned something. Try something else. Over longer periods of time, this process will change your objectives and your strategy.
The best solutions survive, you learn, and the business gets better and more profitable. That’s the point of an analytics program.
Measure it: Manage it!