10 Tips from Google Analytics Pros


10 key takeaways from the Analytics Conference 2014

Last week, Loves Data presented Australia’s third annual Analytics Conference in both Sydney and Melbourne. The event proved to be bigger and better than ever with digital marketing guru, Avinash Kaushik, and Google Analytics Evangelist, Justin Cutroni, at the conference to deliver keynote presentations.

From a day full of insights and learnings, here is our pick of takeaways to keep you thinking:

1. Use Google Tag Manager (or any tag management system)

Google Tag Manager (GTM) allows you to centrally manage your Google Analytics, advertising and other tracking tags across your websites and even mobile apps. It’s a free tool and can help speed up the deployment of new tags based on rules that are defined within the tool.

  • It will make changes and implementing new Tags a lot faster as marketers won’t have to involve the IT department.
  • It’s an easy way to check if Tags are working before they are deployed.
  • Every change is documented and saved, this will help with troubleshooting later on.
  • Each user can have a varying level of access (choose permissions between view, edit and publish)

2. Don’t rely solely on last click attribution

Last click attribution is one of the most widely used forms of reporting because it’s the easiest to explain. It credits the last touchpoint of interaction to selling a product. The issue here is that it bypasses all the other elements that would have played a role in gaining a conversion. Within marketing you would consider a broad marketing mix including social media, email campaigns, search ads, display ads, video campaigns, print ads, flyers and PR work just to name a few avenues. So, giving credit to just one of these avenues based on last touchpoint before a purchase doesn’t sound like a well thought-out strategy. Consider the bigger picture and move past last click.


3. We’re surrounded by data but starved for insights

Data is easy to extract because most of it is automated. What isn’t automated however are insights that are worthwhile and that can push executive actions. Send executives less data and more insights, provide an action plan and spell out business impacts.

Avinash Kaushik recommends spending 15% of resources on capture, 20% on reporting and 65% on analysis for Analytics accounts with at least a year’s worth of data.

4. Think about the way you present data

Executives don’t want to waste time trying to understand difficult graphs. It’s up to data analysts to present the findings in a readable manner. There are many data visualisation tools you can utilise.

Bonus: a lot of them are free. You could try one of these:

  • Give Google Fusion tables a go, you can even create interactive graphs
  • Sunburst data visualisations also provide a neat way to display complex data
  • Data-Driven Documents easily manipulate documents based on data
  • Klipfolio is also a great tool to bring dashboards together, they even offer a free trial

5. With attribution your goal is to be less wrong each time

Attribution modelling is about evaluating what happened yesterday, predicting what will happen tomorrow and then making better business decisions based on these. Customise your attribution model frequently (every few weeks) so it is more refined and moulds around your business. Create a hypothesis and allocate a small budget, test your hypothesis, tweak and revisit it as necessary. And each time you’ll find yourself being less wrong. Reality check: Attribution modelling isn’t a quick fix, it requires time and effort but the results are worthwhile.

6. Apply the 90/10 rule

Spend 90% on people and 10% on technology. Lots of tools are free, including Google Tag Manager, Google Analytics and even third party data. This frees up funding to spend on people, your greatest asset. Justin Cutroni believes in that age-old adage, give a man a fish and feed him for a day, teach a man to fish and feed him for life.

7. Platform, platform, platform

We all know mobile is big, though let’s not forget how multi-faceted a business can be. Marketers need to consider an entire spectrum of platforms from websites and apps to your CRM, call centre and even traditional touchpoints like Point of Sale (POS). Cover online marketing basics by leveraging Universal Analytics to help you measure anything with a digital connection. Begin to test your results and aim for constant improvement then target your audiences to refine your strategy. The message here: always keep in mind the platform, test and learn.

8. Conversation, amplification rate and applause rate

When it comes to social media, keep in mind these metrics: conversation (this is comments or replies per post), amplification rate (how many times each post is shared or retweeted) and applause rate (number of +1s, likes, favourites, etc. per post). Avoid reporting based on number of likes or followers, what you want to see is that people are engaging with your posts and with your brand. Consider the whole customer journey over the long-term rather than obsessing over easy-to-measure metrics.

9. Get your offline data into Google Analytics

The data we get about online interactions in Google Analytics is just a small part of a much bigger picture about your customers’ interactions with your brand. With Universal Analytics, it is now easy to merge in data from offline touchpoints – point-of-sale transactions and so on – to track offline conversions and get deeper insights into customer behaviour.

10. Think people, not sessions or visits

Getting your offline data into Google Analytics is just part of the new way of thinking – better analysis and better insights come from tracking people across their whole journey with you. It’s easy to get caught up on conversions, but what about reaching out to a user’s second purchase, third purchase and beyond. Consider using the user-id feature of Universal Analytics to track your customers across all their interactions with you, on all their devices, to extract unique insights.


At #GAUC2014, we were also delighted to have on board a host of international speakers including Marcia Jung, Product Manager, Google; Chao Cai, Engineering Lead, Conversion Measurement & Attribution, Google; Stefan Schnabl, Product Manager, Google; all from Google US and Vinny Vijeyakumaar from Sparkline Singapore. Representing Google Down Under was Wendy Glasgow, Head of Data Platforms, APAC, Leroy Pinto, Head of Google Analytics Premium, and Peter Evans, Google Analytics Premium Specialist. Loves Data’s very own, Benjamin Mangold, was also on hand to present Universal Analytics insights.

Were you at the 2014 Analytics Conference in Australia?

Was there a presentation at #GAUC2014 that really got you thinking? We’d love to hear your thoughts and ideas for next year in the comments.