4 Best practices for the formulation and implementation of an AI strategy. Part 10 of the series ‘Perspectives on Artificial Intelligence’.

Of all the current technological developments, artificial intelligence is both the most profound and the least understood. We are witnessing impressive new applications, but can hardly foresee their impact on people, organisations and society. In this series of blogs – Perspectives on Artificial Intelligence – we investigate not only the opportunities, but also the intended and unintended consequences.

The huge gap between ambition and execution

Expectations regarding the possibilities of digitisation and AI are high, but what have today’s companies already put into motion? Much less than discussions suggest. The vast majority (87%) of executives agree that digital technologies lead to disruption, but only 44% report that they are sufficiently prepared (source: MIT SMR). The path is long and winding, but there are still plenty of opportunities to get to the destination.

While the need for change is recognised, many companies fail to adjust both their strategy and organisation. Especially when a company is doing well economically, it becomes difficult to question your business model. Why take risks with your existing sources of income?

There is a big difference between market front-runners and followers. Front-runners disrupt themselves; they transform their organisation from one business model to another and benefit from new sources of profitability before they lose their share of the market to newcomers.

Below, we describe a number of success factors for organisations wanting to incorporate AI into their strategy and business model. There is no standard recipe for AI strategy development and implementation (neither is there one for digital transformation). This is therefore more of a best practices overview than a step-by-step playbook.

  1. Do not create scenarios for shrinkage, but for growth

Existing companies have a natural tendency to protect themselves, primarily implementing new technology to safeguard the existing business by saving costs, increasing automation and improving customer service. Many organisations are unclear about the AI-investment business case; relatively few examples and data are currently available. Cost saving is easier to budget for than extra revenue generation from new products and markets.

Nevertheless, companies are well advised to go on the offensive when implementing AI. Research carried out by McKinsey shows retailers reporting sales increases of up to 5% in stores and up to 30% online thanks to AI analyses of customer behaviour, personalised promotions and dynamic pricing. Additionally, in 12 of the 15 sectors surveyed, companies that deploy AI at scale  report profit margins 5 percentage points higher than their competitors (18% versus 13%).

Only those who are able to recognise the opportunities created by AI can make plans for growth. Anyone who prematurely disregards the ideas and business models of new entrants and start-ups as irrelevant is blinkered and does not pay heed to the warning signs that inform us of approaching change.

Scenario analysis is a good way to turn future visions into reality; ‘bringing the future to life’. Contextual or explorative scenarios describe external developments, usually in your own industry, but a broader approach is also possible. These provide insights into how the world around you is changing and how you can respond to ensure future success. Organisations that want to actively shape the future themselves use normative scenarios. These describe the future that you want to achieve and what you need to do to get there.

  1. Design your client’s future

What does your client’s future look like? And how do you identify what this is before your client does? Steve Jobs did not carry out market research because you simply can not ask clients to come up with a product or service that they have never seen before.

This requires a combination of insight and imagination; insight into the needs and motives of your client, and imagination regarding the ways in which you can use AI for problem solving, simplification and reducing costs. Co-creation is a method of getting close to your client. Later in the product development process, communities of clients willing to test your beta versions are of great importance.

Customer journeys are a powerful tool, provided you do not limit these to the existing operating model and customer contact moments. You must dare to start with a clean sheet. The analysis of your client’s future is not about your products and services themselves, but about the results these products and services will bring to your customer. By using this philosophy, Philips Lighting was able to develop a new business model. Rather than bulbs, they sell the function of the bulb, i.e. light. In the former business model, making bulbs that lasted was, in terms of profits, undesirable, but in the new business model this is no longer the case.

  1. Scale innovation through integration instead of isolation

We have all heard stories about organisations who protect the status quo in such a way that they, acting as a type of business antibody, put a halt to all creativity and innovation. To prevent this, new digital initiatives are often developed in seperate departments (often called ‘labs’) or at alternative locations far away from the existing organisation and thereby shielded from the criticism of the established order.

This is an understandable choice, but also a limiting one. The existing organisation and the corporate start-up are not given the opportunity to learn from each other. Even more important is that you do not transform an organisation by keeping innovation at arm’s length. Successful experiments and prototypes need to be scaled up and this will happen more rapidly if they already have a tangible place within the organisation.

A transformation can only succeed if it is implemented company-wide. If there is no integration of a remotely developed business model, splitting the company through divestment or an IPO is the obvious way forward.

  1. Mobilise talent

Formulating and implementing an AI strategy requires competencies which are rarely united in a single person. In-depth knowledge of the technology, insight into the needs and motivations of clients and a strategic overview in order to assess the viability of a business model are a prerequisite. Marketing is usually responsible for customer experience, IT for technology and Finance for strategy.. And you will also need someone who can coordinate the different working and thinking methods of these individuals or departments.

Talent is scarce and the so-called ‘war for talent’ will only intensify in coming years. The good news is that it is becoming increasingly easier to mobilise talent from outside the organisation. For technical dilemmas, for example, you can browse open innovation and crowdsourcing platforms such as NineSigma and Innocentive. Or you might make use of hackathons, which are not only a way to bring in new ideas in a short space of time, but also a method of recruiting new talent. Furthermore, we see that the flexible set-up of temporary employees and freelancers is not only useful for extra manpower during peak periods, but is also a source of specific knowledge and a way to accelerate internal development.

Speeding things up

The impact of AI on a company’s operating model will continue to increase in the coming years. Organisations must simultaneously restructure themselves and continue to sell. You will need to understand new technologies, translate their possibilities into solutions for customers, and design and implement new business models. AI is not an innovation that can be kept to the side-lines. AI touches the company’s core.

Companies such as ING and Philips show that transformation is possible and that incumbents can successfully compete with newcomers. A timely start is necessary, because we often witness a ‘winner takes all’ dynamic within the digital economy, making it (almost) impossible for latecomers to catch up. Execution of the above-mentioned best practices requires, more than anything else, strong leadership.

 

 

 

Comments are closed.