Digital Roundup #132 – Automation, Voice & Analytics

As the new year gets into gear, it’s time to reevaluate what your company has been doing for the last 12 months, and start putting into place strategies that will make 2019 a big one for you. Thankfully we’re here to give you a helping hand with the latest tips, tricks and news from the digital world.


5 rules to win with automated marketing straight from Google

Automated Marketing

With ever increasing demands placed on businesses to keep up with customer expectations, and so many other businesses vying for that all important eye-time, delivering a consistent and effective marketing strategy can feel almost impossible. Thankfully there’s something waiting in the wings to make your life a whole lot easier. Automation.

Powered by machine learning, automation can dramatically reduce the amount of time you need to spend on producing marketing campaigns, freeing you up to focus on the more important aspects of your business growth. But how do you best use automation to drive success?

1. Optimise for growth, not efficiency

According to Google, top performing companies optimise for profitable growth and take a more holistic view to marketing. That’s opposed to those who are preoccupied with increasing efficiency and measuring too granularly.

For instance, focus on long-term profitability or total sales from all your channels as opposed to short-term ROI and online leads. For instance, financial services company Betterment increased its organic brand searches by 245% by engaging YouTube viewers that had searched for financial keywords.

2. Acquire better customers

Given that 20% of your customers drive 80% of your profits, use machine learning to better target your more valuable customers – those with a greater customer life time value (CLV). Google use the example of a travel agency who found that their most valuable customers were the ones booking multiple cheaper trips, not those who bought the most expensive holidays.

3. Get more from current customers

Automation can also be used to increase the CLV of existing customers. You can achieve this by improving cross-selling techniques and lowering your churn rate. By setting a machine to discover what products the customer should buy next, marketing campaigns can be constructed that focus on re-engaging with buyers. Insurance firm Allstate increased customer retention by 2.4x and cross-selling by 4x by launching new cross-selling and churn-reduction campaigns. Find out more here.

4. Work on creative elements

As automation is picked up by more and more of your competitors you’re going to have to do something a little different to stand out from the crowd. Google’s responsive search ads tool can create a heap of tailored ads for a single keyword, generating unique headlines and descriptions tailored to audience groups. Additionally, YouTube’s Director Mix lets businesses quickly create different creative variations for their advertising categories.

5. Improve mobile experiences

With improved advertising campaigns created by machine learning, websites with poor mobile experiences are going to be losing out. The time is now to get your mobile site in shape. Accelerated mobile pages can help slower websites reach a larger audience, and Progressive Web Apps deliver reliable, fast and engaging web experiences.

Getting ready for voice search domination

Voice search

With millions of Amazon Echos finding their way under the Christmas tree this year, we are about to see a huge surge in the amount of voice searches conducted. So if you’re not already, it’s about time that your company got ready to embrace a new form of search functionality. In fact, a recent survey carried out by BrandContent found that users now consider voice search up to par and are planning on utilising voice in the coming 12 months.

What does this mean for you?

Users talk differently to the way that they type. This means the way that they construct their inquiries will incorporate different search terms and keyword combinations. Your perfectly primed pages, ripe for page one will now need to be reevaluated for more conversational language.

Though it doesn’t stop there. With Alexa Skills – specialised apps that allow you to perform certain tasks through Amazon’s voice assistant – you can even start attracting new customers. UKMortgages has created a Skill that helps customers find the right mortgage for them by answering 3 simple questions. Expect to see more of these.

Alexa for Hospitality is already being used by hotels to give guests their own virtual butler, and automotive companies are exploring built-in voice search functions which could answer questions about the vehicle. Expect, too, to see more these industry-specific integrations of voice search.

John Buss, MD for UK and Northern Europe at Yext, explained,

“We’ll see the service evolve to not only provide local dining suggestions, but also book guests a table, secure them a place on the best experience days and packages, and provide an easy way to feed back to the hotel on their service. The possibilities are almost endless.”

Our advice: introduce more conversational language to your best performing organic pages and monitor the results.

How predictive analytics can improve social

Predictive Analytics

Predictive analytics is a rising art being employed by the forerunners of business today, giving industry leaders information on the performance of products and posts before publishing. Saving money, time and – in many cases – face, it can also be used to improve social performance through, “dynamic buyer personas, keywords contextualised about topics and segments, and ads impact prediction.”

But how? Let’s take a little look…

The use of predictive analytics

Perhaps the most primary use of predictive analytics is to extract value from all the data that your business is collecting from all of its channels.

Ester Liquori, CEO at You Are My Guide, suggests thinking of it like Lego bricks,

If bricks are coming from different packs, you have to figure out what to do with them. You will have so many options that you stay puzzled, and you will go ahead by uncertain procedures. This means you are going to spend a lot of time on pieces recognition, mounting, dismounting, trying again.

However, what if they are organised in packs by theme? You know that they are part of a defined shape. Sure, you need to test, but you already know that all the pieces are part of what you want to build.

In this case predictive analytics can be employed to organise the data thematically based on previous actions.

How it gets social

Social media channels are a cacophonous mess of thoughts, feelings and facts. Well, maybe not so many facts. Either way, sifting through constant updates and staying on the edge of conversations is an impossible task, as is separating the fact from the fiction. But with machine learning data can now be sorted into groups based on parameters set automatically by artificial intelligence.

This leads to the creation of dynamic buyer personas. Pooling information on social data means you will have up-to-date profiles perfectly primed for personalised marketing campaigns. Without having to constantly create buyer personas, marketers will be free to focus on the all important creative elements that make a campaign successful.

With a better overview of search and social data you can start to actually contextualise your important keywords. That is, understand why customers are using them – not just what the words are. This better positions you to create copy that captivates your consumers.

Being able to prioritise and segment content and customer data into easy to read chunks, you can make your marketing smarter and cost-effective. Win-win in our book.

How to research, and optimise your website for, questions

Optimising for questions

People are inquisitive. They ask questions all the time, want to find out new things or confirm their suspicions. Here’s where you can come in. By providing answers to their questions you can not only give them what they want, but push them through your sales funnel too.

Writing for Moz – the inbound marketing specialists – Ann Smarty explains that,

  • Niche questions are a powerful content inspiration source.
  • Questions trigger a natural reflex in human beings, making these pages more engaging.
  • Research into industry questions provides a greater understanding of natural language that can be used to optimise for voice search.
  • Answering questions improves organic visibility in Google’s featured snippet section.

What questions to answer

Basic questions

These are your run-of-the-mill questions that have simple answers. How many calories are there in a tangerine? (Answer 53 per 100g.) These should be succinct, correct and easy to understand.

How-to questions

Step-by-step answers. How do I poach an egg? – For instance. Videos can really help here.

Branded questions

These include:

  • High intent questions concerning subjects such as how to buy your products or service.
  • Navigational questions concerning how to access the customer’s account, for example.
  • Competitive questions. Where a user is seeking information on your brand compared to your competitors.
  • Reputational questions – these are concerned with your brand history, culture, etc..

Then, how do you research these all important questions that users are asking?

Google Suggest – the words it appends to what you type into the search box – give you some of the most common inquiries other users ask.

People Also Ask – a series of peripheral searches based on the initial inquiry, POA is a treasure trove of information on user questions.

Discussion boards – the likes of Quora and Reddit are filled with the questions that customers ask. Importantly, the best answers are voted on, so you can frame your response around these.

It seems so simple, doesn’t it? Well, that’s because it is. Just make sure that you’re implementing question and answer research into your content strategy for the following year.


Sebastian Paszek

Marketing manager

Controlling the chaos of the digital landscape, Sebastian is a multiplatform executive, project manager and photographer.