A version of this article also appeared in Forbes on Nov. 18, 2019.
Making the most of AI requires putting it at the heart of operations and changing how your company works
Count yourself among the 96.4% of execs if you believe AI-driven technology is transforming how we do business. This belief is fuelled by estimates about AI’s expected economic impact. A PwC report calculates AI will add $15.7 Trillion to the global economy by 2030.
The relationship between Big Data and AI and our increasing ability to extract value from it using AI underpins these projections.
Data-driven decision making (DDDM) increases Business Intelligence (BI) and boosts competitiveness. However, this requires bringing Data and AI into the core of all aspects of decision making.
When correctly applied AI can transform both the client and employee experience through process automation. But viewing AI as plug-and-play technology with immediate ROI is wrong.
To become AI-driven requires changes to operations. The focus needs to shift away from narrow business problems towards bigger challenges like optimising the customer journey. Becoming AI-driven requires adapting company culture, structures and educating everyone about AI.
For example, AI used in our Automated Lead Scoring tool will enable a business to better identify the leads most likely to convert. But unless the entire sales process is modified to make best use of this information, the benefit is limited.
AI ‘To Do’ list
Firstly, interdisciplinary collaboration and creating cross-functional teams is essential. The biggest impact happens with a mix of skills where operational people work alongside analytics experts and data scientists.
Secondly, adopt DDDM. Augmenting judgement with an algorithm’s recommendations produces better decisions than either a machine or a human can make individually.
These steps involve the biggest cultural change. The traditional top-down decision-making approach is abandoned in favour of recommendations made by neural networks.
Your business isn’t AI-driven if employees need approval from above before acting on the recommendations made by Machine Learning driven tools.
Thirdly, be adaptable. Insisting a business tool has every desired feature before deployment reduces the long-term impact.
Creating AI applications is an iterative process. The test-and-learn approach generates early user feedback which allows correction of minor issues before they become costly problems. Development speeds up and reduces the distance to reaching a minimum viable product (MVP).
It is also essential to identify the unique barriers to change in your business. Some obstacles are common, such as workers’ fear of losing their job to new technology.
However, there are businesses with characteristics that lead to resistance. For example, Relationship Managers are paid to be in tune with their customers’ needs. Some will choose to ignore the tailored recommendations made by an AI-driven tool.
Also, everyone knows managers who measure status on the number of people they oversee. They are likely to object to the decentralised decision making or reduction in reports that AI could allow. Leaders must prepare and equip their workforce to make these changes. But equally important they too must adjust. Most failures are because of a lack of understanding about AI among senior executives.
Learn from other AI-driven companies
Much can be learned from other businesses that have implemented AI. Their success in aligning AI initiatives with their culture can help to guide your efforts.
One financial institution gamified their AI adoption program with a sales contest. Agents who recorded the most conversions using the AI-powered tool were showcased in the CEO’s monthly newsletter.
A telecoms provider implemented an AI-driven CRM at its call centre. They helped employees’ transition to the new software by adapting their approach.
Previously, employees reacted to customers who called to cancel their service. They changed their approach to proactively reach out to customers identified by the CRM to be at risk of defection.
Employees got training and on-the-job coaching on how to close sales. This unified approach reduced customer churn by 10%.
The marriage of AI and CRM solutions is a complimentary partnership. By channelling all business data (phone calls, live chat, emails, chatbots, etc.) to a CRM provides your AI with a central source to mine data from.
Modern CRMs can serve all customer-facing departments (Sales, Marketing, Customer Service etc.) making it ideal for implementing AI-driven practices across the business.
Educate everyone about AI
Successful AI adoption requires everyone being educated about it from the very top down. Firstly, senior execs need a high-level understanding of how AI works. This equips them to identify AI opportunities.
Marrying their industry experience with knowledge about AI will help them estimate the impact on workers’ roles. This also helps them identify potential barriers to adoption and better inform talent development requirements and guide the cultural changes needed.
Secondly, strategic decision makers, like marketers, require high-level training using real business scenarios that show how the tools improve decision making.
Thirdly, for front-line workers a general introduction to the tools with on-the-job training in how to use them.
Expect AI transformation to take months and anticipate a loss of momentum during implementation if results are not immediate.
To counter this, leaders must actively encourage the new ways of working. Early iterations rarely work out exactly as planned.
When that happens, leaders need to highlight what was learned to drive improvements in the next iteration.
Finally, reinforcement of why AI is being used will complete successful implementation. Comparing the results of decisions made with and without AI encourages greater use and acknowledges the employees who drive implementation.
Are you AI ready?
Adopting AI creates unique challenges. However, there are general considerations that are consistent for most businesses.
Audit the existing technology and talent pool within your business to decide if they meet the needs of your AI system. Look at existing data points and your data science expertise. Consider bringing in products and solutions that broaden your ecosystem.
After deciding where AI fits into your business, the gaps in existing tech and expertise that need to be filled will be clearer.
One option is to invest in a startup that has successfully developed their AI tech to augment existing resources. Alternatively, identify vendors to collaborate with or to co-invest in building market-ready applications.
The objective is to take actions that promote scale in AI as this brings its own benefits. This dictates how far the needle gets pushed on BI and the success of AI adoption.