Over the last decade, we’ve seen data become the fuel for organisations globally. It’s spawned business models that could not have existed before, and with that, has forced data to become a market commodity - creating a global information supermarket.
While businesses use it to drive their customer interaction, and - crucially - their sales, harnessing this power has several hurdles to jump. In order to address these, Nimbus Ninety members gathered in the Gherkin and discussed data enlightenment and artificial intelligence (AI) in the context of sales.
AI can’t replace humans; it’s not designed for interaction.
At the centre of sales is the interaction between two people: it’s about asking questions, listening to the answers and together coming to objectives for both sides. Much of the issues businesses face is when they try to replace the skills that salespeople have with AI alternatives. The irony in this is clear: AI has not been designed by salespeople, and therefore, it doesn’t understand the processes and “soft” data (human intuition, contextual information) that a salesperson makes interaction decisions by.
An example of this is in a user visiting and revisiting a website: one member pointed out that such a user is in the “education” phase, and isn’t ready to be sold to yet. Behavioural data is needed here to ensure that the user isn’t scared away with aggressive sales pitches from bots, simply because they’ve visited the site twice. Sales interactions must be nuanced and personalised to each user - just as human interactions must be, if they are to feel organic.
Fundamentally AI brings a different skillset to that of humans - it is vital to understand that it doesn’t simply do human tasks better. Using the skills that AI can bring requires identifying the overlap with human workers and identifying where the strengths of AI outweigh the strengths of human workers (in this case, salespeople). Once these two areas have been identified, these strengths in different loci can be monopolised on and thus drive the business forward. Data shouldn’t be driving the interaction; but rather, the interaction should be driving the data.
One member brought up the need to balance the rational and the emotional when selling. In many organisations, this is a culture change of getting people to trust the data, even if it goes against the preconceptions and hunches they have of a sales interaction. Essentially, salespeople have learnt patterns and AI can help them with that. But it’s also not enough by itself: humans are irrational and don’t always fall into these patterns. Human irrationality is not something that AI can process and quantify: it’s simply a series of outliers within a data set.
AI can’t solve your problems if you don’t understand what your problems are.
When thinking about any technology, it doesn’t work to roll-out application and hope that it will make everything work better. It is crucial to start with the problems you are facing, rather than the technology, and apply the technology to these problems. Often it’s a case of testing the tech-as-problem-solver in a small sample of your customers, and then analysing how well the problem was addressed. This is no different with AI.
Sparking culture change has to start with this attitude to technology as the solution provider for all problems. It is also important to ensure that people aren’t applying advanced AI solutions to basic problems, as well as recognising your buy-in and identifying your current and future competition. Driving this change has multiple parts: training and supporting people to think differently, simplifying processes and putting data at the heart of decisions. Enabling your team to do great things with the data allows them to find the answers themselves - especially when AI cannot give answers, but simply insights.
AI can help with market intelligence: giving insights is its strength.
One member highlighted that data-driven companies can see growth of seven times that of a data-dead incumbent. With anomalies, AI can analyse the patterns and give three or four hypotheses for why those outliers sit where they do. The responsibility then lies with human workers to bring the contextual information in order to reach a judgement on a hypothesis - and take this to sales.
Similarly AI can help with aiding interaction beyond the sale: checking people are using their product to the best of its ability, and truly getting value out of it. Algorithms excel in bringing data together, and allow analysis across data sets. Essentially at the moment, AI gives you the trends from which you can make the decisions: it provides you with predictive analysis.
However, it remains that the people who understand the business will understand the data best: without knowledge of the wider business context, data analysis becomes void of meaning. AI without humans cannot be of any help - but humans without AI are limiting their growth hugely.
This event was in partnership with Tableau, an interactive data visualisation software company.