September 2015

Marketing Analytics Q&A: Exploring Heat Maps and User Experience

If you would like to get these in your inbox weekly, plus our extra commentary, please subscribe on our Connect page!

Discovering Customer Match (1)

What’s the big deal with heat maps? Do they help me understand and improve my user experience? Should we use them on our website?

Source: Eyequant

Source: Eyequant

I’ve got a bit of a love/hate relationship with heat maps. The reason being that while it’s useful to know where people are putting attention on your page, it can also lead you down all sorts of paths of investigation and analysis that don’t resolve anything meaningful.

The product I’ve used several times and introduced clients to is called Crazy Egg, which is co-founded by well-known Analytics expert Neil Patel. Crazy Egg gives you heat maps, scroll maps, click maps and other analysis on your website. For example, a scroll map shows you how far a user goes down the page, and a click map marks the spot where a user clicked their mouse. This data is useful for sure, but often it ‘s hard to actually know what to do with it.

On the other side are predictive analytics tools such as EyeQuant. We subscribe to EyeQuant, and find it useful in analysing designs, in proposals, talking to prospective clients, etc. EyeQuant takes a screenshot of a web page and predicts where the users attention will go in the first few seconds. They achieve this using machine learning algorithms that were trained on 1000s of eye tracking tests.

The other types of related tools capture more than just heat maps. At the moment, I’m looking into tools that record browser sessions, track form abandonment automatically and make the analytics around user experience and conversion much more accessible. The one I’m looking at in detail right now is called HotJar. HotJar boasts heatmaps, visitor recordings, conversions and more all in one product. It’s cool, and probably a little creepy if you’re panicked that some analyst could be watching you browse a companies website weeks later.

The reality is, though, this is needed, especially in highly competitive niches. Qualitative feedback such as reviewing users sessions is necessary to improve design, the user experience and ultimately conversion.

If you’d like us to undertake an analysis of your website with EyeQuant, please get in touch. We are giving away complimentary analysis to the first five readers to contact us!

Google Analytics says my website is a good candidate for Google Analytics Premium. What is it exactly?

Wow! You too? Google Analytics has sent me a notification on nearly every client GA account that I have access to, recommending the same thing. The funny thing is, the notification seems to be general and untargeted as I’ve got the recommendation on “new” accounts without any significant data.

Google Analytics Premium is a significantly upgraded version of Google Analytics. The product, which costs $150,000 USD per year last time I checked, is aimed at Enterprises and those with much bigger data needs than your average business.

It offers higher traffic volume, improved data modelling, integration with Google display advertising tools, real-time data, unsampled data, great query access, and processing priority, etc. If that list doesn’t mean much to you, then you’re probably not a good fit for the product.

The product is certainly relevant to bigger enterprises with millions of visitors and complex data analysis needs. I have used the Google Analytics API before to obtain unsampled data, and it varies positively from the data you access in standard reports. However, at $150,000 per year through authorised resellers, it’s certainly a significant step up from the standard free product that we’ve all become familiar with.

Marketing Analytics Q&A: Designers Who Also Care About Data

Designers Who Also Care About Data

If you would like to get these in your inbox weekly, plus our extra commentary, please subscribe on our Connect page!

We are about to hire someone full time in a digital marketing manager role. Their job will be to help coordinate and execute our digital marketing campaigns, including with some external suppliers. I’ve thought we should get someone with some creative skills as we have a bottleneck right now with an ongoing design. My concern is, they won’t be focused enough on Analytics or being the “guardian” of the performance of our marketing campaigns. How could we handle this?

dataxdesign

I’ve heard this question before, and sometimes the difference between a key team member who is a “creative” VS business minded is seen as a “one or the other” proposition, it doesn’t have to be. Although there is a stereotype that “creative types” are more interested in design and less interested in “business”, the stereotype isn’t always true! For example, I’ve met several creatives who were both business minded and very participative in the analytics and testing process.

When working with diverse teams of this nature, I’ve noticed the most important thing is finding someone who is going to display leadership and take ownership over their role and function. This means practically that they care about what money their company is investing in digital marketing (including their salary), and they’re determined to get a good result. They’re also leaders, so they’re looking to improve their function and relevance in the organisation. Quite often, you can find digital professionals with a niche skill-set such as design, who have the drive and motivation to be a leader, and will want to work with you in a leadership role as they were restricted in a previous job. These types are perfect because they want your organisation to succeed, and they want to have more responsibility and decision making, something that might be lacking if they were locked in under a hierarchical “big agency” structure previously.

To attract these professionals, focus on posting jobs that talk about “Digital Marketing Coordinator” or “Digital Marketing Manager”. Ask about skills (the Adobe suite, such as Photoshop and Illustrator) would be most familiar with designers, and give them situational business questions once you get a shortlist and start interviewing. A professional who knows their way around Photoshop or Illustrator and has the drive to have more decision-making authority is perfect for what you’re looking for.

How did you learn your analytics skills? How can I begin to learn the core skills involved in Marketing Analytics?

In brief, I started running campaigns for clients and various organisations I was involved in as early as 2005. Early on, I learnt all about online advertising, Google AdWords, and banner advertising. I experimented with Google Website Optimiser as early as 2007 and began testing and using Google AdWords to drive conversions for several commercial clients around that time.

I noticed that testing worked. Doing more of what was working and less of what didn’t work led to better results. Strange isn’t it? I started believing in the platforms and looking further into software such as Google Analytics soon after. From these early projects and positive feedback, I began to appreciate the value of testing and using data. I started reading direct marketing books and thinking more about persuasion and what drove people to take action.

After doing this for over ten years, it has all become fairly second nature. More recently, I’ve been experimenting with R and learning newer technologies such as tag managers and personalised content. I’m now putting my attention into identifying the common digital problems that small and medium businesses face, and looking to solve them with solid products and services, of which this weekly newsletter is one of them.

Let’s talk about how you can learn relevant skills. If you’re going to be using Google Analytics for your website, the Analytics Academy by Google is a good starting point. It contains relevant examples about designing a measurement plan, collecting actionable data, and gives you practical examples of the interface. If you’ve got a dual monitor setup, you can have the course notes/videos on one monitor, and your actual Google Analytics on the other.

Analytics Academy – Digital Analytics Fundamentals

Once you’ve got, the essential tools mastered, look into establishing a regular fortnightly process to review the data in your Google Analytics account and begin asking questions about the results. For example:

– What specific metrics do we measure performance by?
– How much did we spend on each channel in the past two weeks?
– What’s our current cost per sale/lead?
– What can I do to improve the results?

I’ve created an excel spreadsheet for you with some examples. Sure, it’s a little boring, and there is software available that will automate the data gathering and put it inside a nice PDF for you to share with your team. That’s not the point, though. Only by analysing and thinking about the data will you get to see improvements. My belief is that if you start doing it manually each week, you’ll need to absorb the data and become familiar with it. If you want to improve your skills, focus on asking “Why has this happened” and look to be proactive in making suggestions for improvement. As you become familiar with the trends and have evidence week to week, you will be able to build up your understanding and begin on improving your website performance.

 Download the Excel Template

Marketing Analytics Q&A: Questions You Need To Ask About Display Advertising

If you would like to get these in your inbox weekly, plus our extra commentary, please subscribe on our Connect page!

Questions You Need To Ask About Display Advertising

What are the most important things to know about display advertising? Who should I buy it from? When should I avoid it?

Your Ad Here (1)

Do you click display ads? I don’t, and according to the global averages, display ads are clicked at a rate of 0.06%. Translated, this means that for every 1000 page views, you won’t even have one ad click. Display advertising is typically sold on a CPM basis, or cost per 1000 impressions. These costs vary from under a dollar, to more than fifty dollars for some websites. So for example, if someone is quoting you 250,000 impressions for 3000 dollars, then you’re effectively paying $12 CPM (There are 250 1000 units in 250k and 3000 divided by 250 is 12).

So what’s good? That depends a bit on your goals, and your tolerance for vanity metrics. If you’re a regular reader of my writing, you’ll know that I don’t have much tolerance for vanity metrics. However, many businesses get excited that their ad got shown 250k times and it’s a good story to tell your boss if they get excited about these sort of things too.

I believe the pecking order should be: Conversions > Clicks > Impressions. In that case, if only put relevance in the clicks and conversions we get, let’s say we get 150 clicks from our 250,000 impressions above (250k x 0.0006), that’s going to be $20 cost per click. 20 bucks per click is probably not going to cut it if we are interested in what people are doing on our website.

So how do we make it work then?

1) We need to use available segmentation criteria to only target people who would actually be interested in our product or service. This should boost up our click through rate significantly. Targeting criteria has improved a lot and will continue to improve.

2) We need to make our creative (banners) stand out amongst everything else.

3) To give ourselves the best chance for success we need to probably pay less per 1000 impressions than $12.

For clients who want to get into display, I usually suggest that they test the waters on Google’s Display Network through Google AdWords. It’s much better to commit $500 to a test than $5000 to a guy like John. If the test works out, then you’ll be better equipped to know what you probably should pay and can negotiate on that basis.

Display certainly has its place. It works well for remarketing and for targeting very specific people with a timely offer. As with all marketing channels, it’s important to measure and test against consistent criteria. Which brings us to question two…

We have just published a new landing page about our new product. The page has a video and a call to action to opt in with an email to get more information. What should we be measuring and why?

I’d measure three things:

1) Total real time spent on the page

I want to know if a visitor spent five seconds, fifteen seconds or longer on my page. Why? As I’m going to be running display advertising, I want to know if these people stay around. I also want to be able to compare marketing channels and have a common interaction goal. For instructions to do this with Google Analytics, refer to the following link:

Tracking Engaged Time in Google Analytics

Once it’s sending in events, we will link those events to goals (e.g. < 10 seconds, 10 to 30 seconds, 30 to 60 seconds, > 60 seconds).

Get in touch with me for instructions for your specific website configuration.

2) Video interaction (plays, pauses, completions)

I want to measure how many users clicked play, and how many actually finished the video. I can then break this down by marketing channel to get a better understanding of actual interaction with my video and landing page by marketing channel.

If you’d like to learn how to do this for your website, get in touch and I’ll send some instructions. I’ve done it for YouTube, Vimeo and others.

3) Contact form submissions

If I’m collecting contact details I want to measure when this form was submitted. Most of the time, this is handled by sending an event to software such as Google Analytics when the submit button is pushed.

Once again, I can link this to a goal and compare results across marketing channels.

Marketing Analytics Q&A: Google Tag Manager and Landing Page Analytics

If you would like to get these in your inbox weekly, plus our extra commentary, please subscribe on our Connect page!

Google Tag Manager and Landing Page Analytics

Tell me more about Google Tag Manager and how it is useful for my website?

Google_Tag_Manager

Imagine a medium to large business, with a complex site and multiple vendors providing support. You’ve got a web development company, a marketing agency, and a web hosting provider. The marketing agency wants to run a new digital marketing campaign with some banners and needs to add some tracking tags and tracking pixels on the website. What used to happen was that the marketing agency would tell the brand manager and the brand manager would tell the web development company. The web developers would charge the client an hour or two to put the tags on, then the marketing agency would need to verify everything was running correctly. More often than not, there would be back and forth and something wouldn’t work properly because someone made a mistake, rinse and repeat.

What tag managers achieve (including Google Tag Manager) is centralised management and control of marketing and analytics tags on a website. Google Tag Manager is pre-configured to run many popular existing tags including Google Analytics, Google AdWords, AdRoll, Crazy Egg etc. You essentially add tags inside the Google Tag Manager interface, and set firing rules for when they should activate. Once you’re happy with your tag configuration, you can test it in a sandbox environment and ensure everything works properly. Once it’s all confirmed, you publish it and that replaces the existing tag configuration on your website until you release a new set.

For developers, it introduces a structured organisation and version control for marketing/software tags, which makes it easier to know what external vendors are running code on the website, and test to make sure things are working properly before deployment live. For marketers, it enables a straightforward way to add tags for common marketing software and speeds up the whole process. For example, in the scenario above, the marketing agency could add the requested tag through Google Tag Manager and it could be approved by the web development company by testing before publishing in the next version.

How useful is it for your website? If you’re just running Google Analytics now, then it’s probably not worth putting on Google Tag Manager right now if you’ve got an existing website. However, if you’re looking to build a new website or run new digital marketing campaigns where you will be using advertising networks such as Facebook and Google AdWords that will provide you with tracking pixels, it’ll be worth putting on to make things easier. I’d say over half the web developers I’ve spoken to, don’t really know how to use Google Tag Manager yet, but that will change as more clients demand it, it’s not too complex once you try it out.

When reviewing the analytics of a landing page, is there one piece of data that you consider most important or revealing? Why?

Landing_Page_Analytics

I like to understand how long users are actually spending on that page, and how that differs between traffic sources. In general, you’d expect that a user would need to spend at least 10-15 seconds on the page to be considered valuable and actually absorb the information being provided. Here is the thing, the way Google Analytics works, it tracks hits as users move between pages. The standard reports that show time on site and time on page are based on the timestamp of the page view when a user loads a new page.

For example, if John user comes to your landing page and leaves without looking at another page, that will be counted as a bounce and time on page and time on site will be shown as 0:00. Even if John spent 5 minutes reading the page, it’ll still show as 0:00 in your reports.

If Mary comes to your landing page and 40 seconds later goes to a second page, the first page will have a 0:40 time on page. If Mary spends two minutes on the second page and then leaves, Google Analytics has no reference for how long Mary was on the second page, so it will record 0:00 on the second page and 40 seconds for the website visit.

Based on this, only looking at time on site or time on page metrics are not as accurate as you probably thought. If anything, they’re probably reporting a whole lot worse than they actually are. If you’re stuck with the default setup, at least set “Source/Medium” as a secondary dimension when looking at the table to see a comparison between your different channels and campaigns. Context is everything when analysing data.

If you want to track with a little more accuracy, you can implement a special code on your key pages that will fire events to Google Analytics at intervals that you specify. You can use these events to understand how long people actually spent on your pages. There are various ways to do this depending on your configuration. Ideally, I like to fire events at 10 seconds, 30 seconds, 60 seconds etc. Once you do this, you can identify how many people spent a certain amount of time on your page and thus have a better understanding of your engagement. You can even get more fancy and track things such as scrolling with external tools.

If you’re technically inclined, or want to send some instructions to your web developer, this is an excellent technical article that explains how such a solution would work: https://medium.com/google-analytics/how-to-track-engaged-time-in-google-analytics-84d9981920da

I want to justify the benefits of undertaking testing to my boss, do you have any examples or real numbers that I can use to help justify the investment?

We do have a detailed case study with some numbers from a real business in an upcoming issue of Digital Success later in the year, however I’ll give you a few practical examples now. When we talk about testing, we’re talking about improving the existing website, marketing channels, and wider marketing funnel performance. Testing is still a relatively new thing to a lot of businesses, and it is more valuable to some more than others. Generally, businesses either directly selling online or collecting customer leads for later can benefit the most financially from selling. Non eCommerce businesses, such as those focused on content can still improve visitor engagement from testing.

As a hypothetical example, a business in the travel space with over 100,000 monthly website visitors spent $50,000 in testing and optimisation. The optimisation agency that they worked with focused on improving the user experience in the booking process by measuring where users were encountering friction and abandoning the process. Once the pain points had been identified, a series of interventions were implemented including changing some of the information requested, and implementing a live chat popup for users that were taking longer than average, plus dynamic remarketing for those users that abandoned.

After testing and the implementation of recommendations, the overall conversion rate improved by 13%. As the business was doing 2.5 million of revenue per month, their income increased by $325,000 per month from the same traffic, generating an ROI on the work of over 6 times in just month! In this example, the benefits of investing were great, and there was evident low hanging fruit to improve.

Obviously, for a business generating $100,000 per month in revenue online, an improvement of 13% would be great, but it would take 3-4 months to pay off the $50,000 of work. Naturally, the cost of doing optimisation internally vs hiring an external expert will change the cost, and it won’t always be 50k. Depending on how much you can improve, it’s usually well worth it. Think about the revenue improvements of improving 10-15% initially on your conversion rate now, and then have a chat to your boss with those numbers!

Marketing Analytics Q&A: Developing a Content Marketing Strategy

If you would like to get these in your inbox weekly, plus our extra commentary, please subscribe on our Connect page!

Developing a Content Marketing Strategy (14)

Hi Neil,

I’d like some feedback on a content marketing strategy for my company. We are in the B2B space, and we are looking to focus some of our content marketing on green/sustainability as our companies main compliance product saves paper vs. the conventional service (Neil: The company has a product for collecting and storing signatures in person for contracts etc. This saves paper and time.). I’m not quite sure where to start, how to target, and how to measure effectiveness. We already have produced quite a lot of content, but we’re not sure how to utilise it in the best way. Here is our URL: <removed>. What are your thoughts?

Thanks,

John

Hi John,

Thanks for reaching out! I’ve checked out your companies website, and I have a top-line understanding of your product. Since your product is in the B2B space, I’m going to assume that your customers (decision makers/buyers) are going to be communicated with on the following channels:

  1. Email
  2. LinkedIn
  3. Twitter
  4. Industry Publications/Websites

unnamed

Since your company is medium sized, I also assume that your sales team has some process to communicate with and close potential customers. This may be a call, in-person meeting, email, or other. To create content marketing that is both practical and efficient, we need to define closely who we are targeting. We may do this through personas/customer profiles that you may or may not already have.

For Example:

  • Your buyers are typically 30-60 years old, occupying mid to high-level compliance/operations positions in organisations. The main benefit of using your product is an operational efficiency gain, with the side benefits of being green, saving paper and being a more responsible corporate citizen.
  • The benefits of your product are clear and definite.
  • Introducing your product to an organisation may present challenges for integration/changing existing internal process.
  • The buyer needs to consider much more than the price of the product to introduce it to the organisation.
  • The buying process is likely to take some time and need sign-off from multiple stakeholders.
  • Being green and sustainable is nice for the companies PR/may align with the companies values.

Based on where your prospects may be in the buying process:

Loiterer – They’re just browsing, and while they have a general interest, they’re not sure that they need a product like yours yet, and they don’t care too much about the specifics yet. (Most prospects are in this category).
Looker – They’re looking for a solution but just gathering some information about options, and they’re interested in learning more about benefits and specifics. They don’t want to be hard sold yet.
Shopper – They’re ready to buy a solution, and they’re assessing specific vendors against their requirements and each other, they’re open to sales presentations.
Buyer – They’re ready to buy from you and want to know how to make a transaction.

Based on this, your content strategy on social media (Twitter & LinkedIn) should be predominately focused on converting Loiterers into Lookers. To achieve this, you want to focus on content that helps the prospect discover the benefits of your product in a wider context (both direct such as efficiency improvements and more comprehensive benefits such as being “green”). You can post content on your company website blog and link to it from social media such as Twitter and LinkedIn. This will also improve your SEO by having more relevant content on your site.

Lookers who are aware they will probably buy a solution in the future, are going to be more likely to subscribe to email further information about your solution with more technical details and specifics based on their type of company/industry. For this, you may have further advanced private content that you send through email on a regular basis such as case studies. They can get on this email list from your website (perhaps from some more advanced content). You shouldn’t hard sell Lookers, but you should focus on providing them value. You might use your email analytics to determine who is engaged with your emails, and reach out to them personally to book in a call or get further information from your sales team after a period.

Once a prospect has moved to the Shopper stage, they should be handled by your sales team and closed through the normal channels.

Throughout the process, Analytics is used to understand:

  1. What content is working, e.g. which social media content brings the most traffic to our website (Google Analytics and Analytics built into the social media platforms such as Twitter and LinkedIn).
  2. What is the market talking about? Keyword research on Twitter or LinkedIn, Google Alerts on the main competitors/industry keywords?
  3. How much will we need to pay to reach the right people beyond our average organic reach to be effective?

Social media platforms give us an opportunity to target our content directly at our target audience. If we know very precisely who we want to reach, we can have a modest budget to reach our target market directly with our content. This applies to LinkedIn, Twitter, Facebook, and other networks. Unfortunately, organic reach is somewhat limited now due to algorithm changes, so just tweeting something and expecting it to take-off is unlikely to work too well.

Using the Analytics of our content marketing, we can understand how much we may expect to pay to get our high quality content in front of our potential buyers and move them on to our own channels with greater attention such as email., By pulling them into a structured communications process as they move through the buying stages, we can structure the process to gain maximum results and improve it with Analytics.

Summary

  • Ensure that you develop customer personas, to target them correctly using content marketing and other digital marketing. If you skip this step, you’ll be communicating with everyone, which is not a focused approach =).
  • Develop content for different stages of the buying process. Recognise that once a prospect is further along the buying process, it’s probably best to engage with them over phone/email with specifics.
  • The segmentation of the customer buying process we are using include Loiterers, Lookers, Shoppers, and Buyers.
  • In the specific example of John’s company, content focused on the benefits of being Green was probably most suited to the wider market to introduce the product category, but is unlikely to be the sole purchasing reason for someone who is weighing up specific options for a solution and comparing vendors. John will need to have additional content and integrate with the website/CRM to appeal to customers at different stages of the buying process.
  • Utilise other communication mediums such as email marketing to speak to prospects that are further along the buying process. Have a smooth transition and a clear process for interested candidates to move along. Integrate the website with an email list and your CRM.
  • Focus on educating and providing value over hard selling most of the time.
  • It’s wise to allocate a modest budget to target your high-quality content to the right audience. Content rarely travels too far organically, but if you’ve got content that is valuable, it’s best to have a budget to put it in front of the right audience.
  • Analytics is used to understand what strategy works and what doesn’t. This includes the specific content you use and the engagement from the audience that you target. Analyse the data and make adjustments to improve.