How Artificial Intelligence might change marketing/PR in the UK

I have been keeping a watching eye on StoryStream for a while now, as they seek to develop their unique, and rather innovative content marketing platform.

Of late I noticed they seem to talking a lot more about Artificial Intelligence in content, so it wasn’t a huge surprise this week when they launched the latest version of their system, which is now powered by Aura, an AI system for brands.

It is early days for Aura, but it clearly has significant potential to disrupt the content management world, which is a space that really could do with some technologically-driven heavy lifting.

You can read more the product here, but basically Aura combines digital asset management, multi-channel publishing and content analytics. The idea being that that marketers can create the best visual content to influence buying behaviour and increase engagement.

To launch Aura StoryStream took the novel approach of creating an event where three key commentators on AI and marketing answered questions from both StoryStream’s CEO Alex Vaidya and a gathered group of customers, friends and AI geeks.

Those taking part included Rupa Ganatra, who is the founder of Millenial 20/20 event which focuses on how brands market to millennials, and in previous years has provided a great showcase for innovative marketing tech startups. Also on the panel was Tom Ollerton who is the Innovation Director at We Are Social, and David Kelnar, who is a partner at MMC Ventures and a key voice from an investor perspective on AI.

The event was fascinating and archived by a video team, so I expect you be able to watch it online at some point in the future. Here though are a few of the key takeaways I took from the event.

1 For AI to work at an optimum level it still needs human interpretation/intervention

To begin with Alex played a piece of music, Daddy’s Car, which is a tune created for Sony using AI that is designed to mimic The Beatles. Personally I loved it as it reminded me of Jellyfish and Pugwash, but it wasn’t to everyone’s taste and is certainly no Hey Jude. Click the link below if you enjoy a bit of power pop.

Alex made the point though that the music had only been created partially by AI and that a French musician had added lyrics and harmonies.

This was to set the theme for the evening, that although there is a lot of hype about AI and its potential. It still requires humans skilled in data capture and interpretation, as well as marketers with creative skills to make the most of it.

For example, there was a lot of discussion about the notion of ‘assisted creativity’ and how AI was a vehicle for enabling humans to spend more time thinking creatively using the analysis that AI had delivered to them.

2 AI is at the top of the hype cycle

Alex mentioned that AI has been described as the new electricity, or the driver of the next industrial revolution. It may yet prove to be both of these things, but for now AI is at the top of the hype cycle. It is likely we will see a lot more cynicism about AI before it begins to be used to solve complicated problems in real world scenarios.

3 Marketers recognise the potential of AI, but seem reluctant to engage with it

I have seen this time and time again in the publishing world. Newspaper companies understand how technology can change and improve their businesses, but they lack the culture that will enable the changes to be made. Tom gave the example of personalised content which he claimed delivered 54% more engagement than traditional content. Yet the percentage of companies experimenting with AI in the advertising and content worlds is tiny.

Rupa added that the companies that were experimenting with AI often saw it as a bit of a novelty on the scale of ‘look at us we have a chatbot handling customer relations - aren’t we smart?’ It will be a while before AI starts to impact on the bottom line of businesses.

Ultimately this set up the VC David to state the obvious — that marketing is ripe for disruption and that AI has huge potential to influence both marketing and advertising. In David’s view they are industries that generate a lot of data and being able to crunch and analyse that data is ideal for machine learning algorithms.

Later on all three of the panelists spoke about the culture at agencies and brands. Paraphrasing what they said — development in AI needs to come from the top down and that almost all C level execs rose to the top because of their skills and experiences in a pre-AI world. So they are reluctant to push for changes and create elements of businesses that sometimes they don’t actually understand.

David also added that for large companies their datasets are one of their most valuable assets and they need to understand them better.

4 There are still risks in using AI, but not what many marketers expect

Tom made the really important point that some marketers simply believe that AI will do everything. He pointed to the recent experiment by Microsoft with its Tay chatbot which began spewing out racist and sexist nonsense within hours of being activated. Had it been limited to certain topics by a human the story might have had a different outcome.

David also suggested that some bigger companies are simply thinking along the lines of getting a bit of AI in. He said that the issues that AI could potentially solve had to be identified first and then AI brought in to help the company to achieve its goals.

5 AI might lead to the end of ‘the big idea’ in advertising.

There was a very interesting discussion about how personalisation of content might kill off the ‘big idea.’ (think Nike’s Just Do It etc) Advertisers and marketers often work on what concept will resonate with the largest number of their target audience.

Personalised content changes that. There then followed a humorous discussion of whether big ideas were that great anyway. For example the number of car company ads which feature people driving on their own in remote locations. So maybe big ideas are overrated.

6 The UK has a booming AI startup scenes — brands and agencies are spoilt for choice when it comes to partners

After Alex asked what should brands and agencies do next if they wanted to experiment with AI? David gave a quick primer on the UK AI scene. He said that brands had two options — there is software created by the mega tech companies like Google and Microsoft, that brands should use. However he acknowledged that while useful bespoke offerings created by one of the 400 or so AI startups in the UK would almost certainly be far more effective.

Once again he underlined that brands and agencies really shouldn’t start with the technology. They need to ask how can machine learning improve business outcomes, and only when that had been considered work with the AI companies.

There is a lot of AI talent in the UK and lot of creative marketers who sense that their businesses could benefit from rationalising some of their work using technology.

How the two groups interact in the next few years will be fascinating to watch.

Pic from Dr Janet Bastiman @Yssybyl