November 2015

Marketing Analytics Q&A: Top Usability Optimisations that Analytics Can Reveal

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

Top usability optimisations that analytics can reveal (1)

When we built our website a few years ago, it worked great on all our devices at that time. Now with new devices with bigger screens and higher resolutions on the market, our web developer is telling us we need to pay $1000s to upgrade our website and be compatible with new standards and devices. Can you help us make an objective decision using Analytics?

Sure. Keep in mind though that new devices will always be released, and while they may not always become bigger in size, there will always be a new feature or other demand. Part of web design standards are aimed to help standardise production and compatibility with new devices. For example, the responsive framework allows the web developer to specify different design specifications and other customisations based on a range of screen sizes.

Assuming that your website was built on a responsive framework, I’d want an itemised list of what they were proposing to change, what devices were not working now, with screenshots or other evidence backing it up. Chrome has a great device rendering tool built in that you can use to see for yourself (go to View Menu > Developer > Javascript Console), select a device and refresh to see how it renders.

Assuming your website renders fine on different devices in the testing tool, you can also use Google Analytics to see if there are any significance differences in conversion or other engagement between devices and screen sizes. You need to select a long enough date range (3-6 months), and look at the following reports:

1) Audience > Mobile > Overview/Devices

This will give you a breakdown of Desktop, Mobile, and Tablet, plus the individual mobile devices. Look at Bounce Rate, Pages/Session, Conversion rate of key goals, etc.

Note: You might expect Mobile to be a bit lower than Desktop for key engagement metrics/conversion rates. Depending on what you’re counting as a conversion, users often take fewer steps on a mobile and do research instead of a purchase/lead. However, a conversion rate of say less than half than desktop may mean that you have some major usability issues. I typically see Mobile VS Desktop conversion rates of 50-80% of Desktop with some outliers either side.

2) Audience > Technology > Browser & OS

In this screen, you can see the differences between browsers. Keep in mind if you’re not using Mobile or Desktop segmentation then some browsers which are common across both mobile and desktop (e.g. Safari or Chrome) will have the data combined which may change the results. You can also click on the Primary Dimension to take a look atScreen Resolution that will give you an idea of the different screen resolutions accessing your website and the engagement of each. It might be that your site just doesn’t work very well on a very common screen resolution that your users are on. In this case, it might be prudent to invest in improving for that particular use case to get some quick wins.

How can I measure website speed using Google Analytics to ensure a fast optimal user experience?

Google Analytics has a website speed tool built in under Behaviour > Site Speed. It doesn’t take many page load samples, so unless you’re looking at a longer period with 100s of samples, beware of a small sample size.

However, a useful report is the breakdown of different pages underPage Timings and also the Speed Suggestions which links to Googles’ excellent Page Speed Insights tool. Often on your website, you’ll have pages that are too heavy regarding download size, or a mismatch of resources loading (e.g. your site will wait on certain things to download, delaying the entire page). If you see obvious problems in this section, it’s best to export the reports and send them to your web developers for further investigation.

The advantage of this approach is that you’ll have some quantitative data to back up that certain pages of your website are slow, and work on fixing that. Around 30% of users will abandon a website if it doesn’t load within 10 seconds, keep that in mind. If slow website speed is a persistent problem, there are further tools that can test and optimise your website.

Marketing Analytics Q&A: The Truth Behind Bounce Rates

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

The Truth Behind Bounce Rates

My homepage bounce rate is over 75%; that seems really high! What can I do about improving this?

There is a quick process to determine if this is a bad thing. Keep in mind; the bounce rate is generally the % of people who just view the page and exit the website without viewing another page (hence the term bounce). However, depending on the configuration of the site, and the presence of further tracking, a high bounce rate may not be as bad as first thought. I’d consider around a 50% bounce rate average, however depending on the traffic profile, this number can be higher or lower.

So let’s talk a bit about your homepage. 75% bounce rate on a homepage would be considered high. I’d want to do two things:

  1. I would want to see the traffic breakdown for the homepage. Are the main people coming to the home page from organic search, direct, or a paid advertising? To do this in Google Analytics you would go to Behaviour > Site Content > All Pages > Click on the Homepage and then select the Secondary Dimension (Source/Medium).
  2. Second, I would want to know based on the traffic profile, is the homepage serving the purpose which it’s intended. For example, is the homepage providing users with a map of where to go next? Does it load quickly without hindrance or major browsers and devices? Does it provide your relevant information upfront promptly and easily?

Often the homepage is slow to load, with lots of different varied content and no clear purpose of where to go next. This poor user experience will contribute to a higher bounce rate as users want to know where to go fast. The other major mistake I often see is people directing paid search or other advertising efforts to their homepage. Always use a landing page and make the content contextually relevant to what the user is looking for. As I said to a client yesterday, “don’t send them into a forest, show them exactly where they need to go, and they are more likely to find what they are looking for.”

I also like to measure things such as real time on page or interactions with major elements (such as watching a video). These are usually triggered by events in Google Analytics and can be setup using Google Tag Manager, or manually in your code if you’re technically inclined.

My bounce rate on Desktop is so much better than Mobile. Why does mobile often have a much higher bounce rates?

In my experience, to find a website that loads quickly with minimal extra content on mobile and gives me exactly what content I want is a rarity. Even though clean, accessible web standards have been around for several years now, it can obviously take time for websites to adopt these and take care to build a mobile experience that provides the user exactly what they need. Designing for mobile is so different to desktop, and while standard responsive web frameworks will collapse everything into a single column, depending on the way the content has been formatted, it can be much harder to find what you’re looking for.

The times where I’ve seen mobile outperform desktop is in specific landing page sequences where the objective was obvious, e.g. “Click this button, fill out this form, and click this next button.” However, knowing the web developers who built this experience, I knew that significant time went to optimise the user experience, including the size of form fields, buttons, and the availability of critical information on a mobile device.

If mobile is key to your business, and you’re getting significant traffic from mobile, but it’s not converting, consider re-designing your mobile experience and QA testing your website on your phone. You’ll be able to identify fairly quickly where the pain points are, and how to bring the most relevant content above the fold. You can also use software such as Hotjar and record some mobile sessions to understand where users are running into trouble and how you may fix it for them!

Marketing Analytics Q&A: Discovering Gmail Ads and its Value to Your Business

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

The Truth Behind Bounce Rates (1)

What’s the important thing to know about Gmail Ads? How do I get started?

Gmail Ads are a new (now generally released) ad format by Google that sits under the Google AdWords umbrella. They are present in Gmail and allow richer integration than standard text or display ads. Gmail ads display based on parameters that advertisers set based on a scan of your email content and other factors known about you such as your age, gender, etc. The big difference between Gmail ads and normal display is that the real estate is prime, and while you may just want to focus on your email and not click an ad, if that ad is so specific and targeted to what you’re thinking about right now, how could you not just click and have a look?

That being said, it’s not completely open to advertisers to exploit. There are a lot of exceptions to what you can advertise in Gmail ads, and it’s far more restrictive than normal search and display. That being said, if your business is not in a restricted category, there are plenty of opportunities to target.

The biggest opportunities appear to be on targeting competitor domain names. You can target people based on the email domains they receive emails from. Great if you want to show customers of a competitor a relevant ad at the time of them reading an email and thinking about the product and service. You can also target by purchase history, job titles, and a whole lot of other criteria.

The creative also allows you to specify HTML, images and text for your ad if your potential customer clicks on it to open it. From what I’ve seen in my research, many advertisers have just started by having clickable banners, but apparently Native advertising (e.g. content with pictures and value add call-to-action to your website) is the fastest way to get conversions and make the advertising pay for itself.

How do I get started? The feature is available now in your Google AdWords account. if you’d like some help setting it up or discuss your specific requirements then I’m happy to have a chat and make some recommendations.

Best of luck!


Can you provide me with a practical example of how Gmail Ads can work for a business?

Barry runs a business selling pet food. Barry has a website, runs some ads, gets some organic traffic and also does offline marketing. Barry believes he has superior products to his three main competitors John, Harry, and James. The thing is, Barry has found it a bit heavy advertising on their business names for Google AdWords and all four are engaged in a race to advertise on each other, hence pushing the bidding price up and wasting money.

Barry implements a Gmail ads campaign to target the domain names of John, Harry and James. He puts their website domains into the targeting tool and selects some other parameters such as country, age, etc. Barry can target customers who have either bought from these other businesses or made inquiries. Since pet food is something that people buy on an ongoing basis, Barry can entice customers to come over with an ad which talks about some tips for taking care of pets and entices the reader to come to the website to learn more.

Barry can also advertise some new pet toys to those customers of his competitors as they are a complimentary product. Barry implements the Gmail campaign and only pays when customers of the other businesses come and learn more about his business. He succeeds because his competitors don’t even know Gmail ads exist!

Marketing Analytics Q&A: Marketing Software Expenses

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

Marketing Software Expenses

Is it necessary to pay for marketing software? Aren’t free ones enough to run my digital marketing?

I didn’t pay for marketing software for a long time, and it’s only in the past 18 months where I’ve been actively seeking out software to help us with our digital marketing and analysis efforts.

There are paid and free digital marketing software that fall into some categories:

– Social media management software
– Web Analytics software
– Research tools and other databases
– Advertising management and SEO software
– Collaboration tools such as project management systems

You can certainly get away with not paying for software, and your website and digital marketing should run just fine. Software becomes useful when you’re exploring optimisation, collaborating with a team, and exploring efficiency improvements.

In my experience, though, if you come across a free software that’s excellent, it’s probably not going to be too long before it becomes paid, or the free trial gets reduced/downgraded once they hook enough users in.

The software which I see as being most useful covers:

– Managing multiple social media accounts from the one place to save time and facilitate approval between multiple users and scheduling posts.  Applications such as Hootsuite, Buffer, etc. help facilitate this.

– Social media reports, including the growth of fans/followers, most popular posts, etc. The weekly summary that comes built into most management software is useful to keep an eye on top line stats.

– Project management software such as Basecamp, Asana, Trello, etc.


What do you spend a month on marketing tools? What have you added in the past six months that we may find useful?

We spend around $700 AUD per month on various marketing support software. Our biggest costs cover private databases that we subscribe to for competitor intelligence and other analysis tools. I think the average company could get away with $150 to $200 per month maximum and still cover a good range of software.

In the past six months we have added:

Crowdfire: Allows boosting and automation of particular social media tasks including identification of new people to follow by categories and keywords and automation, especially on Twitter.
Coschedule: Allows the reuse of content such as blog posts or other social media content at a later time. Allows you to recycle your best content automatically and post directly from blogs which save time.
Basecamp: We had a brief affair with another project management tool, but we came back to Basecamp for its ease of use and integration with our workflow.
Canva: For making images quickly for blog posts without the need for a designer.

What do you use? Anything that you recommend that I can include in my Q&A next week?

Or if you’re using a particular software and you’re wondering if there are alternatives, shoot me an email and let’s talk!.

Marketing Analytics Q&A: Exploring the Analytics Behind Email Marketing

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

Exploring the Analytics Behind Email Marketing

I was discussing email marketing with my friend the other day, and he said that 20% – 30% email open rate is standard. That seems low! Do you have any insights on what’s normal?

Your friend is technically correct, however as with all analytics, context is critical. Email open rate is affected by many factors including:

– List quality/age/frequency of emailing
– Method of measuring open rates
– Industry
– Core messages/subjects/offers

According to Mailchimp, their analysis has given us a breakdown of average open rates/click through rates by industry, You can access that data here: Mailchimp Email Benchmarks

To improve your open rates and effectiveness, focus on doing the following:

  1. Always adding new emails to your list. Emails which have been on your list for a long time are likely to become stale eventually.
  2. Test subject lines between campaigns. Aim for subjects which are catchy and entice the reader to open to find out more.
  3. Show up consistently. Email every week, or every fortnight, or every month. If you just email ad-hoc, you won’t build up an expectation of you showing up.
  4. Potentially segment your list and send them specific offers/emails. For example, you could segment by location, purchase type, gender, etc. You might need to collect/merge extra data to do this.


Our emails often have clicks to our website, but we don’t know how to measure the visitors past that. Can you give us some tips?

It’s great that your emails get clicks to your website, but then what? What’s also important is what pages the users visit, if they purchase, contact you, etc. So how do you measure this?

The standard and best way if you use Google Analytics is to use UTM tracking. UTM tracking is a method of telling Google Analytics a little bit extra about a visitor when they arrive at your website. In the case of email marketing, it involves adding UTM tags as additional parameters at the end of your URLs that users click on. Some email software packages also add UTM parameters on automatically, or as an extra option in the settings.

To track your next email, I suggest you utilise UTM tracking parameters. There are three core parts you will be adding on to your URL:

utm_campaign – This refers to the name of the campaign, this might be the name of your campaign, e.g. “October Newsletter”.
utm_source – This is the referral source that the visitor came from – e.g. “MailChimp” or “buyerslist”.
utm_medium – This is the marketing channel that the visitor came from, in this case, it would be “email”.

Google has a tool to build UTM links which you can access here: Google URL Builder

How do you use this data?

This data ends up in the Acquisition > Campaigns section of Google Analytics. This allows you to view engagement on a per email basis, and for advanced users, allows you to segment your analytics data by users who just originated from that particular campaign. You can also see conversions by that campaign if they are defined in Google Analytics.

This is useful because you can then actually measure what those 30 or 100 users wanted to find on your website.