How machine learning is changing the face of business communication

It has long been said that an organisation’s most important asset is its people. Employees define a company, shape its culture, and often prove to be its most valuable source of inspiration, ideas, and ‘shop floor’ insights.

In recent years, this mindset has changed. Now, in the era of digital transformation, it is believed an organisation’s biggest asset is not its people, but rather its data. Data is what shapes products and services, improves customer experiences, and ultimately defines brands.

To downplay the role of people, is to fundamentally misunderstand the way that businesses work and the way that ideas are collaboratively formed. The fact is that tomorrow’s organisations will be defined by both people and data.

>See also: AI: the next level of smart customer service?

Big data has successfully provided businesses with a wealth of information on how their staff choose to interact, when people log in and clock off, and when and where people may be at their most productive. What it does not yet do, is capture the vast quantities of unstructured data that exists throughout an organisation’s workforce; those watercooler conversations and flashes of inspiration that don’t necessarily happen within the quantifiable forums of email and instant messaging.

These are the human aspects of office communication that the average data platform cannot yet capture and turn into meaningful insights. This is soon set to change.

Insight through communication

As more and more businesses adopt voice and video conferencing systems to manage their internal and external communications, a growing wealth of data has been made available, which increasingly captures workplace discussions, thought processes, and worker preferences. These face-to-face interactions can tell us so much more about how and why business decisions are being made, providing a greater degree of insight than any employee survey or corporate feedback form.

Given the unstructured nature of these insights, it is considered impossible for businesses to capture them in a useful, yet non-invasive, way.

>See also: Harnessing machine learning to drive B2B relationships

The idea of automating such analysis is viewed as impossible, with automated review processes simply unable to cope with the levels of nuance required to identify patterns in complex video and voice content. At the same time, the idea of manually trawling through such content to find patterns was both unrealistic and, frankly, unethical.

But now, a third way is starting to emerge; a way that allows businesses to develop valuable insights from unstructured data, without impacting corporate or employee privacy.

Anonymous machine learning

Over the last 12 months, machine learning and artificial intelligence has dominated much of the conversation around how technology will shape the way people live and work in 2018 and beyond. Some real-world applications have come to the forefront, but much of this discussion still revolves around future-gazing predictions, theoretical concepts, and hype.

Although society is still be a long way off from the robot uprising predicted by many AI pundits, it is starting to see some of the most interesting and promising applications for machine learning tech.

Over the next two years, machine learning will become increasingly integrated within workplace communications systems, helping to analyse complex patterns in user data and, ultimately, improving productivity across each interaction.

Already, Fuze – for example – is working with data scientists to prototype such an idea, using a small sample of internal communications data to uncover seemingly inconsequential patterns in how users prefer to communicate and interact in order to improve the overall productivity of the company.

>See also: How Tesco is using AI to gain customer insight

Currently, this analysis relies on nothing more than simple ‘breadcrumbs,’ looking at factors such as whether users typically prefer to start video calls with their camera turned off -data that can then be used to help define default settings within the application itself.

This may seem like an inconsequential improvement, through the introduction of machine learning and big data analysis, but Fuze plans to ultimately expand this process to include all variety of workplace communications styles and preferences.

Spread across Fuze’s 250 million identities around the world, these preferences could then be used to build a detailed understanding of how employees and customers like to communicate. From a business’ perspective, this provides vital insights to help teams increase productivity and become more profitable in the ways they work.

Selective sales and marketing

As an example, consider the potential for a sales representative or contact centre employee. By using machine learning patterns, a sales rep will know when and where to tailor their communications approach–knowing how much they should talk versus how much they should listen.

Similarly, patterns identified could also help teams understand which customers are the most likely to respond to voice versus video calls, along with which times of the day sales reps are most likely to receive a positive interaction.

>See also: The role of artificial intelligence in cyber security

This is clearly a great opportunity for businesses to increase sales, but it is also just as much of an opportunity to provide the best possible customer experience. If a business can use such patterns to streamline interactions and minimise customer frustration, then the result is a win-win scenario for everyone involved.

And it’s not just customers who benefit. Communication is a deeply personal aspect of business, with different customers and employees having their own personal preferences and expectations. Machine learning allows organisations to customise their approach, tailoring office communication methods to each individual’s’ unique ‘personal workflow.’

The ROI of office communication

As it stands, Fuze’s work within this field remains only a localised, internal trial. That said, as machine learning technology evolves, we expect to see such trials quickly evolve into a new standard for enterprise collaboration apps.

By relying on machine learning to analyse the available data, workplace communication providers will ensure that the data being analysed remains anonymous and outside the view of human eyes.

As this closed, automated environment becomes a reality, businesses are expected to become increasingly open to the idea of sharing their internal communications data in this way, allowing their apps to provide more detailed analysis into the ways they work. As this data set grows, businesses may one day reach a point where they can finally put an ROI figure on all workplace communications.

>See also: What is machine learning?

The idea of being able to calculate that a call was only 45% productive and then identify that, through basic tweaks, the following week’s call was 20 percent more productive, is now rapidly becoming a reality. Similarly, the ability for boards, managers, and sales teams to use this data to calculate a firm ROI for each interaction is not far behind.

This is where machine learning will sit in the future of workplace communications. People will always be a brand’s most valuable asset – collaborating, being creative, and defining the big ideas. Machines will streamline this process, allowing businesses to understand how those ideas will be shaped, repeated, and executed at scale.

Sourced by Michael Affronti, VP of Product, Fuze

Browse

Article by channel:

Read more articles tagged: