The age of artificial intelligence (AI) is upon us. In the past few years, vast improvements have been made in how well computers can recognise objects in images and understand human voices.
Progress in these areas has been made due to increased computing power and the availability of large stores of data, which, when combined, have made AI systems dramatically more effective.
These same forces are also being used in marketing. AI, or ‘cognitive’, marketing systems use industrial computing power, big data, and machine learning to improve marketing performance.
While cognitive marketing has not yet been deployed to a great extent, it soon will be. According to IDC, more than half of all companies will be using cognitive marketing by 2020.
So, what exactly is cognitive marketing and how will brands use it?
To find out, Econsultancy, in association with IBM Watson Marketing, recently held roundtable discussions in Delhi. There, senior client-side marketers discussed the impact of cognitive marketing on brand messaging and how they see the technology developing.
Below is a summary of the three main ways marketers on the day plan to use cognitive marketing.
1. Segment audiences in new ways
Segmenting audiences is a key part of providing relevant messaging to consumers. Participants noted that most marketers use demographics to segment their audience into groups with similar wants and needs.
In contrast, cognitive marketing systems search massive data sets from a wide variety of sources, such as web analytics, social media, and purchasing behaviour, to find customer segments which exhibit similar behaviour.
In some cases these segments may resemble traditional demographic groups, but in others cognitive marketing may find common behavioural characteristics among people who appear to be very different.
What this means for marketers, according to attendees, is that cognitive marketing will transform the customer list into a database where each member is connected to others in many different ways. The result is that one customer will be part of countless segments depending on their observed behaviour.
So, a woman aged 34 would no longer be simply considered as ‘female, 30-35’ but, instead, she would be a ‘fashion lover’ who ‘takes three months to buy’, ‘travels to Bangalore twice a week’, and ‘tends to open emails on Tuesday’.
Without using cognitive marketing, one participant noted, these sorts of segments would be nearly impossible to build, manage, and use effectively.
2. Personalise content
With these behavioural segments, marketers can use cognitive marketing systems to personalise content more effectively than ever before.
After receiving core content, a cognitive marketing engine could redesign the messaging so that virtually every person saw something different. Participants envisioned that the system would use data from social media, browsing behaviour, and even sentiment from customer service communications to reformat content for an individual.
Attendees offered a couple of reasons why brands will use this approach. First off, cognitive marketing would ensure that the brand message was delivered in the right way at the right time for each customer. Long-form, engaging content could be sent when you know a customer is at home and shorter, easy-to-consume messaging would appear when they are commuting, for example.
Personalisation would also ensure that brands avoid delivering irrelevant messages and risk being ‘tuned out’ by the customer. Put another way, one marketer said, ‘you have five seconds to get their attention with something relevant, otherwise you are done’.
Following the event, Antonia Edmunds, business leader at IBM Watson Marketing, had a few more words to say on this topic:
3. Help customers make better decisions
Delivering a personalised message at the right time benefits the brand for the reasons mentioned above, but delegates noted that cognitive marketing will also help the customer make better decisions.
As cognitive marketing can make inferences using data from a wide variety of sources, it can also help brands identify customers who have a particular unstated need, said one attendee. This will allow the brand to deliver personalised offers and guidance.
For example, if someone is price sensitive at the moment, they could be told about a lower-priced product range. If they are time poor, the brand could let them know about a new convenience. And if they are in the middle of a major life event, say moving or getting married, the brand could offer to help them with the process.
In this way, according to participants, cognitive marketing will help companies start conversations with consumers around topics which matter to them and not just around what the brand wants to say about itself.
This, in turn, will demonstrate that the brand anticipates a consumer’s wants and needs and, ideally, make it easier for them to choose the brand above all others.
A word of thanks…
Econsultancy would like to thank all of the marketers who participated on the day and our table leaders:
- Antonia Edmunds, Business Leader – IBM Watson Marketing.
- Gowri Arun, GBS Marketing Leader – IBM India/South Asia.
- Joseph Sundar, Business Development Executive, ISA/ASEAN – IBM Watson Marketing.
- Harsh Anand, CSP Leader – IBM Commerce.
We hope to see you all at future Delhi Econsultancy events!