Artifical intelligence (AI) has been the talk of the town over the past few years. Elon Musk has predicted a Skynet scenario unless AI is properly restrained. Facebook has implemented AI to detect suicides before they occur, and AI is the backbone behind Telsa’s self-driving cars.
What exactly is AI?
According to Pedro Domingos (Author of Master Algorithm) – “AI is a type of computer science that can reason and perform common sense tasks. This includes learning, reasoning, planning, interpretation and language-processing. Intelligent design means AI is dynamic (unlike traditional programming) and able to behave differently based on the data provided.
What is machine learning?
Machine learning is a subset of AI that allows computers to learn at an incredibly fast rate, without requiring extra programming. Machine learning allows computers to learn from and make predictions based on data.
Savvy marketers are taking advantage of this new frontier and reaping the benefits. Let’s look at some 7 use cases of AI (and machine-learning) in digital marketing.
1) More efficient Customer Service
Text-based customer service was one of the first uses for AI in digital marketing. Facebook messegner chatbots are being adopted en masse by many companies. It’s likely you’ve already dealt with one without knowing.
2) Optimise Product Pricing
Pricing is one of the most crucial parts of the marketing mix. Yet things can get a little complicated when your product is in different parts of the world . Local factors such as tax, foreign exchange rate and your positioning can greatly affect your margins. Cue AI – this allows dynamic pricing optimisation – which uses machine learning to calculate the optimum price based on sales volumes over time. It also has the potential to recommend which content and language would perform best in certain geographical regions. Pretty cool huh?
3) Personalised Recommendations
We all hate generic content that is not tailored to our needs or interests. Many companies have started working with IBM’s Watson to allow a ‘cognification’ of their digital assets. This can help combine user data and third party data to come up with relevant recommendations. When Netflix ‘knows’ what you want to watch next? That’s AI in action.
4) Image Recognition
Google’s next big play. This will allow you to interpret images – so all you will need to say is “that photo of me at the beach when I was in Bali”. Marketers will be able to exploit this new technology for image searching (think SEO but for images) and greater personalisation for the customer.
5) Enhanced Targeting of Ads
Programmatic advertising (where brands use real-time bidding to buy inventory off publishers) will continue to evolve. As Andrew Ng, Chief Scientist at Baidu Research, reports, ‘Machine learning [is] able to handle more signal for better detection of trends in user behavior. Serving ads is basically running a recommendation engine, which machine learning does well.’
‘Machine learning [is] able to handle more signal for better detection of trends in user behavior. Serving ads is basically running a recommendation engine, which machine learning does well.’
This relevance of targeting will lead to greater conversion, thereby reducing Cost Per Acquisition. Everyone wins! Machine learning will further be able to recommend what ad copy is performing best, and to what particular demographic.
6) Automation of rules-based systems
I’m all for shortcuts if it leads to a similar or better outcome. The typical role of a marketer includes:
- Determining the optimal channel to reach customers
- What is the right message/positioning the customers respond best to?
- When is the right time to target the customers?
This is usually completed by a rules-based systems where defined events can trigger certain actions. For example in drip marketing, if a customer watches a video for longer than 40 seconds then a particular type of content is displayed.
AI has the potential to answer and automate all these questions. Machine learning is able to analyse behaviour and make recommendations in real-time. This may also eliminate the need to run thousands of A/B tests to figure out the best content for each segment. Predictive lead scoring is another function that can be handled by AI. Imagine knowing in real-time, how likely a visitor to your website will end up purchasing your product?!
Will this mean the end of marketers?! Of course AI will not replace the marketers, rather just the repetitive and mechanical tasks. This will allow marketers to focus on:
* Building creative campaigns,
* Delving deep to understand customers behaviour
* Figuring out core problems
* Define positioning
AI will be like a marketer’s virtual assistant!
7) Improved acquisition from emails
It’s easy to see machine-learning within email marketing platforms. Marketers will be able to combine first party and third party data into their campaigns. This will allow AI to bolster customer profiling, leading to ‘personalisation on scale’.
Marketers will be able to combine first party and third party data into their campaigns. This will allow AI to bolster customer profiling, leading to ‘personalisation on scale’.
AI (such as Einstein) can already predict which content is likely to provide the best conversion rate, to what demographic and the ideal time to send off a campaign.
With so much potential, the future of AI and machine learning is exciting. We are only just beginning to scratch the surface now… soon AI will be ubiqitouts throughout the digital marketing world – Elon Musk, watch out!