Webová analytika – proč? kdo? kdy? jak? (Insight Model)

transcript

Google Confidential and Proprietary 1

Web Analytics: From Data to DecisionsWhy do it? Who owns it? When is it important? How?

Johannes OppermannAccount Strategist & Google Analytics PoweruserGoogle Austria

Client Logo

Conference on Web Analytics –Web Statistics and Their Evaluation

May 28th 2009

Agenda

1 Why you should care about Web Analytics

2 Who owns the Web Analytics function in your company?

3 When should you do Web Analytics?

4 How can you go about implementing a Web Analytics Process?

Why you should care about Web Analytics

Am I creating effective content?

How do I improve site interaction?

How can I make my marketing campaigns more

effective & accountable?

How are users engagingwith my site?

You are facing difficult questions about your Site

Where and why are visitors abandoning my

shopping cart?

Google Confidential and Proprietary

What does “Web Analytics” mean, anyway?

The Official WAA Definition:

Measurement, collection, analysis and reportingof Internet data for the purposes ofunderstanding and optimizing Web usage.

Or, much shorter:Are you successful with your site or not?

Useful answers & actions for everyone

• Track ROI on all online initiatives from a single report

• Identify long-term trends in online marketing programs Executives

• Get the most conversions at the lowest possible cost

• Target high return markets and market segments

• Test advertising creative for each of your campaigns

Marketers

• Demonstrate the value of your site usability improvements

• Show how your site design decisions improve website ROI

• Identify web design parameters like visitor connection speeds and browser types

Webmasters

Analyze all marketing campaigns

Determine which marketing initiatives are driving the best traffic to

your site

Banner Advertising

Search Marketing

Email Marketing

SEO, Referrals, Affiliate Programs & Offline

You get to know what happens on your site

Web analytics tools give you CONTEXT and visibility into the full on-site USER EXPERIENCE

Who owns the Web Analytics function in your company?

Web Analytics should be owned by the business

Is your company big? Then it's probably less agile.

Avoid taking only the HiPPO (Highest Paid Person's Opinion) into account. Ask the users!

Web data can be complex. So you need someone smart to interpret it.

Also, Web Analytics can't be owned by the IT department alone. Because they're solving other problems.

It should be owned by whoever will be made responsible if the website fails. Marketing? Sales? E-Commerce?

When should you do Web Analytics?

We'll relaunch next week. Have some numbers?

Web Analytics isn't a one-off project. It's something that should be integrated in your company's processes.

Analysis

Insight

Decision

Action

New Data

How can you go about implementing a Web Analytics Process?

Focus on the bottom line

People will listen more to you if you make a difference in their lives: Your stakeholders will support you if you can help them get a larger bonus.

So, the question is: are you focusing on outcomes?

Example: "We make 30% of our sales on the web. A conversion rate of 5% instead of 3% will mean 100.000 € more revenue."

Now we're talking.

It's not about your data. It's about your decisions.

Numbers are meaningless without context. Make them objective.

Insight = Data + Analysis.

Q1 2008Q2 2008Q3 2008Q4 2008

wethe others

Q1 2008

Create a data driven culture in your company

Analysis

Insight

Decision

Action

Outcomes

Clickstream Analysis• Unique visitors• Top Landing Pages• Bounce rates• Conversion rates

Outcomes Analysis• Support – problem resolution• Commerce – revenue • Lead generation – conversion

rates• Could your users do what they

came to do?

Choose the right metrics to solve for

Questions you can ask yourself• How much do we sell / convert? (Outcomes)• What are users really looking for here? What's happening? (Behaviour)• Why are the users clicking on random things? (A/B-Testing, Labs)

Actionable insights and metrics that drive change

Insights – analysis – actions

Clickstream analysis: We're only inferring intent.

Try to find trends and patterns, then segment & drill deeper.

Once you're done, formulate a hypothesis and test it

Example: "I think our signup form is too long. So conversion rates should go up if we leave out all the unnessecary fields."

Experience Analysis: • Testing• Experiments• Usability studies

Necessities & Tricks of the trade

• Decision making – "It's not my opinion, it's the data!"

• Give context – data over time, other competitors, user tests.

• Listen to customers, Listen to bottom line

• Bet on outcomes ("I think page A will convert better!")

• Reporting is not analysis – reports will not tell you what to do by themselves. So go exploring! At least 20% unstructured analysis.

• Empower analysts, encourage a culture of risk taking (Failing means learning what to do better)

• Test and try things out!

It's all about continuity

Website Optimizer

How can Google contribute? Google Analytics.

Use Google Analytics to:• Understand how visitors engage

with your site• Compare performance of all

marketing efforts from emails to keywords to offline campaigns

• See all your AdWords Campaign data, and track non-Google campaigns as well

• Identify areas for site improvement• Pinpoint breakdowns in conversion

processes• Track e-commerce metrics, such

as revenue, average order value, and conversion rates

Free, hosted web analytics tool that helps advertisers improve their sitesand increase their marketing ROI with 80+ standard reports.

Our own A/B Test with the Picasa download page

We measured outcomes by comparing the number of downloads each of the landing pages generated.

Google Confidential and Proprietary23

A: Uses “Free”, Picture, Action Oriented

Google Confidential and Proprietary24

B Page Increased Downloads by 30%!

B: Cleaner, Button, Value Proposition

Thank You!Q&A

Where to find more information

Google Analytics Website: www.google.com/analytics

• Create an account, find implementation instructions & FAQs

Official Google Analytics Blog: http://analytics.blogspot.com/

• Get updates on new features & tips from Google practitioners

Website Optimizer Website: www.google.com/websiteoptimizer

• Learn how to get started with A/B and multivariate testing

Occam’s Razor by Avinash Kaushik: www.kaushik.net

• Blog by Google’s “Analytics Evangelist” provides insightful posts for web analysts and marketers of all levels of experience

"Web Analytics an Hour a Day" by Avinash Kaushik (Wiley Publishing, heavily used for this presentation)

Webová analytika – proč? kdo? kdy? jak? (Insight Model)

Documents