December 2, 2013 | Monica Guy

Digital Analytics: A Better Measurement Plan

analytics

Remember the days when content decisions were mostly made according to the boss’s whim? Well – they’re still here for many companies. But not for long. It’s now so easy to see what customers think and feel about your content, you would be crazy to ignore it.

This is where content marketing gets FUN. With a simple measurement plan in place, you can:

calculate the ROI of different marketing channels and campaigns
identify issues which are harming growth and conversions
improve the user experience – leading to higher conversion rates and increased loyalty
test different marketing campaigns or versions of e-newsletters and ads for effectiveness
identify areas of possible growth, online and offline
maximize organic search through more targeted keyword usage
identify and maximize high-quality traffic from referring sites
refine target demographics and identify new markets
track growth in the community of users
learn more about what customers want in order to sharpen future marketing campaigns
attribute conversions to specific marketing campaigns

So some of your strategies might crash and burn. Who cares? Failing fast means improving fast and with real-time data there’s no need to wait for quarterly reports to figure out what works.

Measure twice, cut once

But stop. Before you go fiddling with your analytics program you need to prepare the ground with a solid measurement plan. Google Analytics processes data on the spot. If you don’t configure it to give you the data you need, you can’t go back and reprocess the data. It won’t be there.

What follows is an improvement on the measurement plan recommended by Google in their recent Google Analytics 1.01 course. Steps 1, 3, 4, 6 and 8 made up their 5-part process. I find steps 2, 5 and 7 to be immensely helpful too.

9 steps to a meaningful measurement plan

Document your business objectives
Document your online presence
Identify your business strategies and tactics
Identify key performance indicators
Identify your testing and measurement toolkit
Decide how to segment the data
Identify contexts
Choose targets
Identify a relevant attribution model

Here’s what it means

1. Document your business objectives

What is the ultimate purpose of your business or organization (not your website)? Do you want to sell shoes? Raise awareness about a social issue? This defines the entire measurement strategy.

2. Document your online presence

Having a comprehensive overview of your company’s online presence will help you to identify strategies, tactics and priorities right from the start.

How many websites do you control and how are they organized?
Do you have a blog? An app?
Is any of your content multilingual?
Which social media platforms are you on, and how many followers do you currently have?
Do you have a presence on external sites such as TripAdvisor?

Note that this is not the same as documenting your technical infrastructure, which is another prerequisite to a successful analytics tracking program.

3. Identify strategies and tactics

Strategies are the methods you use to achieve your business objectives. Your online presence may employ any or all of these common strategies:

E-commerce (sell products or services)
Lead generation (collect potential leads)
Content publishing (encourage engagement and frequent visitation)
Online information (help users find information)
Branding (drive awareness and loyalty)

Tactics are the methods employed to achieve these strategies. For instance, a “branding” tactic could be to use a particular colour theme in all communications. A “lead generation” tactic could be to offer a free download in return for the user filling out a lead capture form.

4. Identify key performance indicators

Key performance indicators are the yardsticks used to measure the success of a company’s tactics, strategies and ultimately business objectives.

E-commerce KPIs will be heavily focused on sales – the number of sales, value of each sale, loyalty of customers and so on. Content publishing KPIs are more about reach, awareness and user engagement.

5. Identify your testing and measurement toolkit

Remember you’re not limited to Google Analytics. Free tools for gathering competitive intelligence include Alexa, Technorati and Google Trends. You might want to conduct in-depth experiments, in which case a paid service such as SiteSpect will be handy.

You should also identify methods of gathering qualitative user feedback such as surveys and usability tests. Analytics can tell you a lot, but you should let customers speak for themselves whenever you can.

6. Decide how to segment the data

Segments allow you to isolate and analyze subsets of data. Segments could be:

date and time (user performance on different days of the week or times of day)
device (user performance on desktops, tablets, mobiles)
marketing channel (effectiveness of different marketing channels and campaigns)
geography (which cities, regions or countries perform the best and where the opportunities lie)
customer characteristics (first-time customers, repeat customers and so on)

Not segmenting data means you’re looking at everything in aggregate – making it impossible to see what’s happening at the micro level. Get rid of the noise by choosing the segments that are relevant to your KPIs and focus on those in order to get meaningful insights.

7. Identify contexts

Contexts can be external, such as industry benchmark data or a competitor review.

They can also be internal. For instance, you can measure your website’s current performance against its historical performance. Different departments can measure their performance in relation to each other (and hopefully share some best practices too).

8. Choose targets

Choose realistic targets for each KPI, taking into account the contexts identified in the previous step.

For instance, an e-commerce target might be to increase the average value of each sale by 10 percent over 12 months. A branding target might be to increase followers on social media platforms by 30 percent over three months.

9. Identify a relevant attribution model

This is the one we all care about – the ROI of different marketing channels and how to assign value to each.

The standard “last click” attribution model assigns all value to the last interaction before a customer is converted. So if a customer interacts with the company on social media, browses the website, disappears for a week and then clicks on a display advert and makes a purchase, that advert gets 100 percent of the credit.

More sophisticated attribution models recognize that the customer’s prior interactions “assisted” in the conversion process. The one that probably most matches reality for e-commerce businesses is the “time decay” model, in which the customer’s interactions just before the conversion receive most of the credit for the conversion, but previous interactions are given a proportion of the credit for their role in assisting the process.

If your KPI is around branding or awareness, the “first click” model is most appropriate as this marks the moment in which the customer first interacted with your brand.

Monica G. – Content Strategist