Welcome to the MediaGamma blog


The gap is widening between those with a handle on their AI strategy and those only scratching the surface of its potential.

6 artificial intelligence business opportunities in 2020


Posted by Victor Malachard

AI business opportunities

6 artificial intelligence business opportunities in 2020

Most business owners and decision-makers already understand the growing value of artificial intelligence (AI). AI helps brands keep up with the competition, improve business processes and deliver more value to customers. A recent global study found that while the majority of companies are still in the early stages of adopting AI, most have begun fast-tracking AI business applications. 

Whether you run a small business or a large corporation, you should consider harnessing the power of AI. But even in the midst of a scramble to rollout AI solutions, there is still widespread confusion about the nature of AI, and the real ways in which it can change business processes. The first step to success with AI is picking the right project and accurately setting your expectations. 

To that end, we’ll look at six key AI business opportunities you might consider investing in this year. We’ll also create practical frameworks for how to get started with each.

 

1. Chatbots


Chatbots have become an integral part of eCommerce, and are one of the most popular forms of AI. In fact, it’s estimated that more than 80% of online customer interactions will be completed without humans this year.

Chatbots use algorithms to understand and respond to customer questions in real-time. In some cases, chatbots are now nearly indistinguishable from human customer service representatives. Although this is still a controversial point, the chatbot ‘Eugene Goostman’ first ‘passed’ The Turing Test in 2014. Although there is still a way to go before chatbots regularly convince users that they are ‘human’, the technology has developed a great amount.  

Using chatbots can help you reduce service time, make fewer mistakes and increase engagement. You can also deploy chatbots as a data collection tool. IBM estimates that business will save $8 billion by 2022 through the use of chatbots.     

 

How to get started

The simplest way to get started with a chatbot is to use a chatbot creation platform, which allows you to bypass the need to program anything yourself. To build a capable chatbot, you’ll first need to define your goals. You’ll also need to think about how to integrate the bot into your website’s backend. Once you deploy your bot, you’ll need to optimise and reevaluate regularly to ensure continued success.

As you build your chatbot, you should consider what data you’d like to gather from your customers. You can then customise the conversation flow to be most effective at gathering this information. However, if you’re using your chatbot to collect data, you’ll need to consider the impact on customer experience, along with regulatory concerns, including GDPR. Like with all data collection decisions, you should take your choices seriously. 

 

2. Business intelligence

Decision-making matters. To maintain a competitive advantage, it has become increasingly critical to harness the power of data-driven decision making (DDDM). DDDM can result in increased revenue, better customer experiences and greater operational efficiency.

To assist with smart decision-making and problem-solving, more businesses have started using Business Intelligence (BI) solutions. In essence, BI uses real-time data to help you make efficient and effective decisions. The most simplistic BI solutions are just dashboards, making it easy to review data. However, AI algorithms can be deployed to sort through that information and help you focus on what’s important. Using BI software can help you ask and answer data-related questions through analytics, reporting tools and data visualisation.

 

How to get started

You don’t necessarily need an in-house team of IT specialists to get started with BI. There are plenty of user-friendly, self-service software options, such as Google Data Studio, that let you collect information and view it in a dashboard. However, some businesses find it challenging to integrate a BI system with their current system and generate the right reports that lead to the most actionable insights. If this is the case, you might consider seeking advice from an AI consultancy.

Whether you’re going the DIY route or seeking outside expertise, you should always start with defining your goals, including KPIs and metrics. You should also think about the questions you want your data to answer and the actionable insights you hope to glean. Keeping a goal-oriented mindset will help you wield your BI tools to their highest potential.

 

3. Data analytics 

The terms ‘BI’ and ‘data analytics’ are often used interchangeably, but there are differences. While BI focuses on the presentation of historical data, data analytics focuses on making predictions. Basically, while BI concentrates on the “what”, data analytics tends to place more emphasis on the “why”. Data analytics deploys big data, data mining, statistical analysis and predictive analytics to determine future trends. It can also establish complex or unobvious relationships between data. If you need to generate reports based on marketplace trends or create future forecasts, data analytics is the way to go.

Data analysis can help your business retain existing customers and acquire new ones, gain valuable marketing insights, mitigate and manage risks, drive innovation, and assist with supply chain management. Fully-fledged AI systems aren’t the only data analytics options. However, the complexity and volumes of data involved make it an ideal application. Worldwide data is predicted to grow 61% to “175 zettabytes” by 2025 — there are huge business opportunities in turning those raw figures into insights.

 

How to get started

Big data analytics is performed using advanced software. These technological tools are agile and work quickly and efficiently to complete analysis, often focusing on specific industries and verticals. Nozzle, for example, is an innovative AI-powered solution centred on eCommerce data analysis. By looking into SaaS products in your industry, it’s very possible to find a customisable or off-the-shelf solution able to supercharge your current operations.    

Although SaaS might be the simplest solution, more customisation can be achieved using in-house teams or working with consultants. You could even take a middle-ground approach, customising an existing product to fit your exact needs. If investing heavily in data analytics, bringing on board at least a few specialists can help you maintain, hone and optimise your system, even if you get outside help to set it up. 

 

4. Behaviour prediction

In the past, determining your customers’ wants and needs involved a fair amount of guesswork. However, sophisticated technology can now predict and explore the ways people behave. Although really a subset of predictive analytics, behaviour prediction (the dedicated focus of data analytics to predict human behaviour) has such powerful potential that it’s worth a unique mention.  

The emergence of deep learning has made predictive behaviour analysis much more accessible, and accurate. Deep learning is a machine learning process that uses “deep” neural networks to learn skills and solve complex problems. With large data sets and significant computational power, deep learning can predict human behaviour with reasonably high accuracy. 

Through deep learning, behavioural prediction tools can analyse vast data sets to help you predict customer behaviour. Relevant information might include preferences, shopping habits or cart abandonment rates. The goal is to paint a clear picture of what your customers will do next so that you can be prepared to meet their needs and potentially influence their future behaviour.

 

How to get started

Good news: if you already have large swaths of data that you’ve been gathering about past, present and potential customers, deep learning allows you to put this data to use to predict behaviour.

Another bit of good news: Behaviour prediction technology is becoming increasingly accessible. For instance, Google offers cloud software that allows you to build and host a deep learning project. Although expertise isn’t necessary to get started, it can help you get more out of your tools. Depending on the scope of your project and your specific needs, you might consider bringing on a consultant.

 

5. Programmatic

If your business advertises online, programmatic can help you save time and more effectively target your intended audience. Programmatic ad buying uses intelligent software to purchase online advertisements. The software allows you to bypass manual orders on platforms such as Google or Facebook by using an automated, auction-based process. Automating the process frees up human resources to focus on optimising and improving the actual ads. 

Many brands find that programmatic not only saves time and human resources, but it also saves money and improves sales performance. The opportunities and ROI of programmatic investments will only increase as digital channels become more important to brands. In 2019, 65% of digital ad spend was already being done using programmatic tools in 2019.   

 

How to get started

Again, if you’ve already been collecting data about your customers, you’ve got a head start. This data can help you determine the customers you’d like to reach. You can also use your existing data to set goals and build a strategy. Once you’ve covered these basics, you will use a platform to access an  auction to bid for your target audience’s views. 

When exploring options in programmatic, you’ll notice that you have the choice of fully managed services, self-service software tools or hybrid solutions. Some popular software options include Beeswax, SmartyAds, TubeMogul, and Simpli.fi

One thing to remember about ML/AI tools is that they will improve to your specific use case the longer they have access to your data and processes. By experimenting with your exact data sets, the AI can learn what success looks like and alter its processes to increase a good outcome.  This is one reason that programmatic is such a good long-term investment. It might have upfront costs, but you not only achieve efficiency savings (that will add up over time), the solution will grow and improve within your specific environment. Lastly, it’s worth noting that the marginal cost will decrease as the business scales. As always, if you’re not confident in your abilities or simply need help getting started, a consultancy can help.



6. AI consulting and machine learning-as-a-service (MLasS)

MLaaS and consultancies aren’t exactly an AI application opportunity. Rather, they are growing business opportunities for the deployment of AI — helping you bridge the gap between theory and action with shocking speed. 

The MLaaS market is predicted to reach a valuation of $8.315 billion by 2023, and there is a good reason. By combining the simplicity of cloud hosted SaaS with machine learning capabilities, companies like IBM, AWS, Google and Microsoft are delivering basic ML capabilities on-demand. These tools can provide classification, regression, clustering, detection and recommendation capabilities that in-house developers can use to either support bespoke projects or provide off-the-shelf functionality.  

Although machine learning and AI are not identical terms, it is a powerful and practical subset of software solutions that are able to use data to “learn” for themselves. By delivering this as a cloud-hosted package, the barrier to entry is dramatically reduced. However, keep in mind that ML is not just about running models. You need to ensure that data cleaning, feature engineering, evaluation and monitoring are all done correctly.

The growth in AI consultancies is similarly fueling the simplification of AI adoption. Hiring in-house staff isn’t always the best (or even a possible) option. Particularly in an emerging and competitive market, getting your hands on the best people is expensive. For small business, there isn’t always the need for that kind of expertise long-term, even if it’s needed to get things started. 

PwC predicts that AI will contribute up to $15.7 trillion to the global economy by 2030. Consultancies can help any company find the most effective way to participate in that growth. This might be helping you set up and configure a MLaaS-based solution. Or it could mean building your own tool from scratch suited to your exact business needs. Fundamentally, you will get help evaluating your data and picking your best option.       

 

How to get started

Both of these trends and opportunities are themselves a way to get started. The main task is to start investigating some specifics. Look into the capabilities on offer by the main MLaaS providers, and then look into consultancies near you. 

Many MLaaS vendors provide APIs and platforms so that you can run basic operations even if you don’t have private infrastructure or expertise in data science. Service providers also offer data storage. If you don’t require storage, most MLaaS platforms can support third-party data storage.

A big thing to think about when looking into consultants is their ability to help you execute a project. A large dividing line in the industry is between consultancies with operational capabilities, and those only really able to advise you on investments. Particularly if you don’t have a lot of in-house expertise, looking for help bringing in talent and executing the rollout of your project can have a dramatic return on investment. 

 

Using AI to deliver results

The value of AI is unquestionable. But many people still have questions about how best to deploy it within their business. As you’ve probably noticed, the first steps in getting started with AI usually involve goal-setting and looking at existing capabilities. It’s important to consider the problems you hope AI can solve, and the data you have available to tackle those problems. This will help you decide how viable an AI project is, and help you decide how you want to use AI to achieve your goals and improve your business. 

Fortunately, whatever your needs, there’s probably an AI-powered solution that can help. And whether you’re looking for short-term assistance in launching an AI application or a longer-term strategic partnership, consultants can help you deliver on your goals faster and more effectively. Remember: AI is no longer a tool of the future. It is a present-day necessity.

 

Interested in finding out how AI can deliver results in your business? Leave us a message about your current situation and we'd be happy to help.

Get in touch

Victor Malachard
Victor Malachard
Executive Chairman

Related Posts

How to decide on your business’ first AI project: 5 steps

Scientists and their discoveries have led to great revelations throughout history and, in recent years, none have proven more impactful than the discovery and development of artificial intelligence (AI). First coined by John McCarthy as early as 1955, delivering on the promise of AI has been a goal of scientists ever since.

AI in-house talent vs consultancies: what's right for your business?

Businesses across sectors are scrambling to capitalise on the value of AI right now, with usage growing by as much as 270% in the last four years alone. Despite this influx, it can be hard to get started here, or to know which of the many arising AI implementations is best. Still, with predictions widely pointing to the fact that 80% of technologies will have AI foundations by next year, getting your head around the issue should be a primary concern. 

6 examples of Artificial Intelligence changing business

At the start of 2020, an impressive 37% of organisations were already utilising AI systems in some form. That’s a 270% increase in the last four years alone, and it’s hardly surprising considering available worldwide big data is set to grow by as much as 61% by 2025. In other words, as the need for unique data handling and simplified processes comes to the fore, companies who don’t get on top with AI soon stand to drown in the influx.