AI-Driven Growth Best Practice Guide

Artificial Intelligence (AI) is currently one of the biggest trends in marketing, with the prominence of tools like ChatGPT making its appeal more mainstream and accessible.

While it may feel like a recent phenomenon, with developments increasing at an exponential rate, the technology has actually been around for some time. It’s a key capability already baked into many existing tech tools - CDPs, MAPs and DAMs, for example. Many other core martech tools use AI to automate and enhance functionality.

Advances in larger more accessible data sets and data processing have allowed AI to take hold in recent years.

With increased data access via the Internet of Things (IoT) making our devices and their data increasingly interconnected, combined with AI applications, we will only see accelerated growth. So while mobile, internet and cloud technology all made huge advances, some argue that AI will make the biggest paradigm shift we’ve ever seen, evolving much faster to dwarf the impact of previous technologies.

As costs reduce, AI technology is becoming more accessible, as well as more mainstream.

AI represents a huge opportunity with a limited number of organisations currently using it to facilitate real-time business decisions and improve customer experiences. Studies show that 63% of B2B marketers are not using AI in their tech stack.

In this report, we’ll explore the growth and evolution of AI, examining some of its potential uses, especially in marketing, sales, and customer experience.

AI definitions and current trends

AI as we know it now is the result of years of research and experimentation, to the point where it can now deliver results for marketing and sales. For this reason, the market for AI technology is growing rapidly.

In this section, we’ll look at the evolution of AI, current trends, and the investments made into this technology by the tech giants.

What is AI?

Artificial intelligence (AI) is the simulation of human intelligence by machines and computer systems to enable problem-solving.

There are different types of AI. Narrow AI (also referred to as Weak AI or Artificial Narrow Intelligence (ANI)) is focused on performing specific tasks.

Narrow AI is the most commonly used, powering many of the applications currently used by businesses. Examples include Siri, Alexa, chatbots, and the recommendation engines used by Spotify and Netflix.

Artificial General Intelligence (AGI) is a theoretical form of AI where a machine would have an intelligence equal to humans. Artificial Super Intelligence (ASI) would surpass the intelligence and ability of the human brain.

Both AGI and ASI are currently theoretical. A fictional example would be HAL, the computer which controls the spaceship in 2001: A Space Odyssey. Its malfunctions also reflect the
potential downsides of AGI.

Sam Altman, co-founder and CEO of OpenAI, the company responsible for ChatGPT and DALL-E, sees Artificial General Intelligence (AGI) as the next major technological breakthrough.

For a full glossary of AI terminology, please view the Glossary section at the end of this blog.

History and evolution of AI

AI, and the idea of machines that are able to think like humans dates back to ancient Greece.

In 1950 Alan Turing published a landmark paper where he speculated on the possibility of machines that ‘think’. As a result, he noted that the idea of ‘thinking’ is difficult to define, so he devised the Turing Test, a test of a machine’s ability to exhibit intelligent behaviour equivalent to, or indistinguishable from, a human.

The term artificial intelligence was coined by John McCarthy in 1956, and the technology continued to develop through the next few decades, as computers gradually became more powerful.

One of the key problems holding back the development of AI until the 1990s was the lack of computational power required to exhibit intelligence.

During this period, landmarks in the development of AI were reached, such as the defeat of chess grandmaster Gary Kasparov by IBM’s Deep Blue computer. More recently, Google’s DeepMind AlphaGo artificial intelligence defeated the world’s number one Go player Ke Jie.

More recently, Meta AI announced the development of Cicero, which it claims is the first AI to achieve human-level performance in the strategic board game Diplomacy.

Diplomacy is different to mastering a game such as chess, as it hints towards advancement in other skills, namely social. Players must show empathy during the game, use natural language, and build relationships to win.

Thanks to Moore’s Law, which states that computing power doubles every two years, AI has continued to grow significantly. Now we’re in the age of big data, in which we collect quantities of data that would be impossible to process manually. AI makes the ingestion, management, and use of such vast data sets possible.

AI now exists as a feature in many types of technology - virtual assistants, autonomous cars, and many marketing, sales and business tools.

Are we entering the AI age?

The development of AI, and the capabilities offered by the technology, means we may be about to enter a transformational period. We’ve had the labour economy and the creator economy, and now we may be about to enter a conductor economy driven by AI, whereby the orchestration of AI technology becomes the driving force.

AI has the potential to change business, far beyond sales and marketing, but also the way we live. Right now, it has the potential to improve the effectiveness of marketing and sales activities, creating greater efficiency, and helping to use customer data to create better customer experiences.

The huge sums of money now spent on AI investment and research means the age of artificial intelligence is inevitable.

In the long term, super-intelligent AI will have capabilities to solve problems far beyond our own intelligence limits, solving current issues around energy, famine and health.

Many concepts that were purely theoretical only decades ago have now been bought to fruition, thanks to advancements in many fields.

The decades of hitting technological walls are now over. But this had to come as a result of improvements in a number of industries: engineering, biology, psychology, mathematics, linguistics, and even experimental philosophy. AI is a real tech collaborative effort.

Each of these offers humanity an undeniable chance to move towards artificial general intelligence.

Sarah O’Neill, Digital Content Producer, LXA

AI trends and data

Data shows that, while many marketers are yet to fully adopt AI, those that use it are already seeing benefits.

• Top-performing companies are more than twice as likely to be using AI for marketing (28% vs 12%) according to Adobe’s Digital Intelligence Briefing.
• The AI industry could be worth more than $15T by 2030.
• 64% of B2B marketers consider AI valuable for their sales and marketing strategy.
• 84% of digital marketing leaders believe using AI/ML enhances the marketing function’s ability to deliver real-time, personalised experiences to customers.

There are obstacles to the greater adoption of AI, and the utilisation of AI features within current sales and marketing technology platforms. These are around the lack of knowledge and skills required, as well as some ethical concerns.

• Marketers say the top obstacle to integrating AI is applying it in their current role and workflow (33%)
• Only 13% of B2B marketers are very confident with their knowledge of AI, while 55% are somewhat confident, and 33% are not at all confident.
• 65% of companies are unable to explain how their AI systems make decisions and predictions
• 78% of enterprises are poorly equipped to consider the ethical implications of using artificial intelligence.

LXA’s State of Martech 2022/2023 report found that 60% of marketers were planning to use AI to drive marketing strategies in 2023.

Market trends around AI

Investment in AI reflects the current potential of the technology.

Perhaps most notably, Microsoft has been investing in OpenAI, the company behind ChatGPT and DALL-E, since 2019. Microsoft invested $1 billion in 2019, and has followed this with a deal said to be worth $10 billion over several years.

The partnership allows OpenAI to access the funding and cloud-computing power required to crunch massive volumes of data and run the increasingly complex models behind DALL-E and ChatGPT.

For Microsoft, the technology can potentially provide the edge as it seeks to compete with Alphabet (Google), Amazon, and Meta in this fast-growing space.

Alphabet (Google) is said to have invested more than $120 billion into artificial intelligence over the past decade, including the acquisition of DeepMind in 2014.

With Meta and Amazon also making big investments in AI technology, startups in this space have been attracting attention recently.

AI will be pushed into public use and consciousness through investment and the presence of these tech giants. Interest from these companies almost guarantees growth.

• AI content platform Jasper.ai received $125m in a funding round which valued the company at $1.5 billion.(6)
• IBM has acquired several AI companies over the last three years, including Databand.ai, Turbonomic, ReaQta, MyInvenio and WDG Automation.
• Google acquired AI startup Alter for $100m, and is also looking to invest in Cohere, another AI startup.

There will be more serious investment in the AI space. 2023 will be the year of AI-driven innovation across marketing, sales and CX.

AI has been baked into many powerful martech and salestech tools for some time but the advance in capabilities of AI-specific platforms is what should be watched.

Carlos Doughty, CEO, Course Instructor, LXA

AI algorithms decoded

AI drives the algorithms powering some of the most successful businesses. These algorithms are designed to surface interesting and relevant content to keep users watching, listening and scrolling.

• TikTok. This addictive social app uses an AI-powered recommendation engine to hook users in and keep them engaged. It seems to depend less on view and follower counts, and more on understanding user preferences to deliver relevant content to keep audiences on the app for as long as possible.
• Netflix. The AI algorithm at Netflix is built around the relevance of content, and ensuring that people can find movies or shows to watch. Netflix recognised the problems around choice paralysis which can occur when users have thousands of viewing options. Using customer behaviour, it ensures that users have plenty of relevant recommendations to keep them busy and reduce churn. It also employs AI in areas such as thumbnail selection. Netflix AI generates thumbnails by annotating and ranking hundreds of frames to determine which thumbnails are most likely to attract a click from users.
• Spotify. Spotify faces the same potential problems around choice paralysis when users have the choice of the majority of music ever recorded. Spotify uses a machine learning tool called the approximate nearest-neighbour search algorithm to group songs and users together based on shared attributes or qualities. AI is used to analyse audio and extrapolate key aspects, such as the energy or ‘danceability’ of a song. It then uses machine learning to assess the effectiveness of recommendations in order to improve over time.

AI use cases in sales and marketing

AI technology is baked into many existing sales and marketing technology tools. Indeed, many vendors boast of AI-powered insights, optimisation, and so on. This means that AI may be already accessible within your current tech stack. It may be a feature you’re already using or a capability that you’ve yet to make use of.

In this section, we’ll look at use cases for AI across various areas of sales, marketing and CX, and the tech tools and platforms which allow you to use AI.

Custom AI platforms

AI-specific platforms are emerging offering a range of use cases for sales and marketing.

• Pencil. An AI ad generator that allows brands and agencies to quickly create variations of ads. This can be used for the generation of ad creatives, and
  automated optimisation of creatives.
• Albert.ai. This is a cloud-based AI platform that plugs into a digital advertiser’s existing paid search and social accounts to optimise performance.
• Varos. This AI solution helps marketers to benchmark performance against competitors, informing buying decisions, and enabling more accurate reporting of performance.
• Drift. Using AI, Drift sifts through emails from subscribers and takes the appropriate action, routing emails to appropriate departments, and freeing up resources.

Generative AI

Generative AI brings artificial intelligence into a new area. It offers the potential to create new content from scratch, whether this is text, images or video content.

Text is currently the most advanced area, with OpenAI’s ChatGPT tool currently the hot topic among marketers. Natural language is difficult to get right, and current AI models are better at shorter pieces of content. However, as models improve, we can expect to see higher-quality outputs.

Tools such as ChatGPT have improved rapidly though, so it may not be long before we see quality longer-form content. If it can pass MBA exams, we may be almost there.

The Generative AI landscape has grown rapidly, and now includes a number of apps capable of generating content for marketing purposes. Some of these tools are costly, but many are affordable for businesses of all sizes.

In addition to text, Generative AI can be used for:

• Code generation. This has the potential to increase developer productivity. Tools such as GitHub CoPilot can autocomplete chunks of code, repetitive sections of code, and entire methods and functions. One benefit is to make the creative use of code more accessible to non-developers.
• Images. DALL-E is the most famous example of tools which can generate images from text inputs.
• Speech synthesis has been used in products such as Siri for some time, but enterprise applications are improving rapidly, with text-to-speech tools now available.
• Video. Tools such as Synthesia allow anyone to create high-quality videos using a text-to-video function.

Generative AI in content marketing

While generative AI may move into longer forms of content in the near future, it currently has some very obvious uses for content marketing, including:

• Planning blog post topics
• Keyword research
• Content personalisation
• Generation of headlines and meta-descriptions
• Generating product descriptions
• Scheduling and creating social posts
• Generating SEO-optimised content

AI has many potential uses in terms of search. If AI generated content is able to achieve prominent rankings, as well as adapting to changes in search algorithms, it could disrupt the practice of SEO as we know it.

On the other side, AI will be used to power search results. Microsoft is already planning to incorporate ChatGPT technology into its Bing search results, which could enhance its natural language search capabilities, and potentially produce more accurate results.

In order to succeed in this new world, marketers will have to adapt in the following ways:

• Search algorithms will increasingly be built around user intent and intent-driven actions rather than keywords or links. Instead of focusing on keywords and links as ranking factors, marketers will need to focus on providing users with the right
  information at the right time in order to drive conversion or engagement.
• The new face of search marketing will rely heavily on AI technology, which means marketers need to develop their own AI-based tools in order to stay competitive (and relevant).

Now that generating content is easy, the question becomes… is the content good? We run the risk of a content race-to-the-bottom, where the internet is riddled with high volumes of low-quality content. Not so different from now, only this time it’s generated by a large language model.

The question you need to be asking is not, “How can I generate more content?” No, that’s a question with an easy answer – log into ChatGPT and have at it.

The question you need to be asking is, “How can I generate more GOOD content?” That problem is a lot harder – and will pay more dividends in the long run. The GOOD content problem has solutions (shameless plug for Phrasee time) – and these solutions drive real impact.

Parry Malm, CEO, Phrasee

Personalisation

As customers demand better, more relevant experiences, the challenge for businesses is to deliver personalisation at scale. In practice, this means personalising every customer touchpoint based on the immediate context and previously observed behavioural data.

Used within customer data platforms (CDPs), AI can add another layer of personalisation to first-party data collected from users. Rather than basing personalisation solely on historical customer data, AI can allow marketers to personalise based on probable future actions.

AI use cases for personalisation include:

• Improved customer retention. Rather than waiting to send churned customers  ‘win-back’ emails, Ai can help predict customers likely to churn and intervene before they do.
• Product recommendation. Machine learning algorithms allow marketers to move away from rules-based recommendation systems to more intelligent and relevant outcomes.
• Conversion prediction. Rather than retargeting ads to all prospects, AI can help you to focus ad spend on those with a greater likelihood of converting.
• In-app recommendations. Predictive modelling allows you to show customers products and offers that have a high likelihood of resulting in purchases.
• Dynamic decisioning. Use propensity scoring to present offers such as coupons and discounts where it will have the most impact, not on customers who are likely to convert anyway.
• AI-powered chatbots. Personalised AI chatbots have the ability to gain deeper insights from users than traditional website forms.
• Customer sentiment analysis. AI can help better identify the true sentiment of users rather than generalising from customer interactions. By learning individual users’ personality quirks, AI systems can assess when users are satisfied.

Channel journey orchestration

Understanding customer journeys plays a key role in delivering excellent customer
experience With so many customer interactions to analyse across multiple channels and touchpoints, this is a challenge for marketers.

AI and machine learning can be used to analyse and orchestrate optimised customer journeys at scale. AI can ingest customer signals from all channels, update the customer profile, and then propose the next best actions that are relevant to those customers.

Every piece of data about a customer triggers AI to maintain the relevance of messaging. For example, an abandoned purchase could trigger a text or some other message. These message-based next-best actions are an important part of the customer experience. AI-powered decisions can also inform the creation of web pages and mobile experiences tailored to the customer based on the updated information in the customer’s profile.

While traditional customer journeys are mapped out in straight lines, the difference with AI-driven journey orchestration is that it remains customer-centric, and next0-best actions are driven by that specific customer’s behaviour and profile. AI-guided journeys are able to be more complex, including many variables, which are all scored according to propensity.

Workflow and communication automation

AI in workflow automation is another big trend in the enterprise. AI-powered automation lets businesses draw on data patterns and machine learning to employ predictive analysis and insights for improved processes. 25% of companies already use AI in workflow automation, while 51% of enterprises are planning to do so in the near future.

AI workflow automation simplifies and improves existing business processes that are often manual and human-dependant. It allows for the encoding of more processes into systems with AI and automation, which can then be efficiently performed autonomously.

This automation can reduce human errors and eliminate many time-consuming and repetitive tasks, such as manual data entry. By adding automation, businesses can create improved capacity for scalability.

Workflow automation can:

• Create processes that reduce time and costs
• Streamline task management
• Minimise errors from manual entries or oversights
• Automate approval and document flows

Workflow automation can be applied to most types of workflows. For example, you can set and automate approval workflows for all of your team members, or automate the employee onboarding process.

AI use cases for workflow automation in marketing and sales include:

• Marketing operations teams use workflow automation for marketing campaigns,
  customer communication channels and for measurement of metrics and marketing analysis.
• In sales, CRM software also provides workflow management. It automates customer communication, form completion and collaboration between teams. For example, a CRM can automate required workflows when a customer has taken a specific step, such as downloading content.

Data ingestion and management

AI can enable the full scope of the data management lifecycle, from ingestion to curation to discovery, and analyse data sets that would take humans years or longer.

Data Ingestion is a process of acquiring and integrating information from different sources into a single repository, typically a database, data warehouse, or data lake.

Once the data is in the central repository, it can be accessed and analysed by teams or by analytics and related systems that look at all the data to extract useful themes and insights.

For example, a customer data platform (CDP) ingests data from marketing automation, CRM, ERP, web analytics, social media, and other systems before cleansing the data through resolving identities, deduplicating profiles, resolving discrepancies between data, and discarding inaccurate data.

This cleansed data is then available to use in analytics and can be delivered to external systems that need it for campaigns and programs.

The use of machine learning and AI in data ingestion can automate and streamline what would otherwise be time-consuming processes, ensuring that data is collected and cleansed accurately.

Testing and optimisation

AI can help companies to speed up testing and optimisation by automating key processes and interpreting results. It can optimise the customer experience at scale in real-time, improving conversion rates and overall performance.

It can be used in areas such as digital commerce and email marketing to create more relevant customer experiences.

AI use cases for testing and optimisation:

• Digital commerce site search. AI can harness product intelligence and customer intelligence to deliver a personalised experience when users search for products.
  AI-powered site search can combine rich product data with shopper profiles based on preferences and behaviour to enhance search results and optimise conversions.
• Product recommendations. Shown on product pages, key landing pages and emails, more accurate product recommendations, based on shopper behaviour and purchase history can increase conversion rates.
• Email subject lines. Subject lines can make or break email campaigns, and AI tools such as Phrasee help marketers generate and optimise email subject lines. AI tools can learn your brand voice through the data you use to teach it, then draft
  subject lines that increase opens and earn higher revenue.
• Optimising send times. Usually, marketers use methods like A/B testing to determine the best times to send emails. AI-powered tools can use historical data to break your marketing campaign into hundreds of sub-campaigns internally, then maximise inbox delivery by sending more targeted emails.

Conversational AI

Conversational AI includes technologies such as virtual assistants and chatbots, that can communicate with customers. These applications are often used for customer service, with many companies deploying them on web, mobile and social platforms.

Conversational AI applications are often used in customer service. They can be found on websites, online stores, and social media channels. AI technology can effectively speed up and streamline answering and routing customer inquiries.

Using machine learning and natural language processing (NLP), conversational AI learns from interactions and should improve the more it interacts with customers.

AI chatbots can deliver a number of benefits.

• Reduced costs. A live customer service interaction, whether by phone, email or web chat is more than $7 for a B2C company and $13 for a B2B company. Chatbots can help to reduce general queries and ensure that only calls that need
  to be handled by live agents are directed to them.(14)
• Service quality. By filtering out repetitive queries, chatbots enable human agents to concentrate more on complex issues which can help improve customer experience.
• Personalised customer service. Chatbots can also use customer data to give more relevant and personalised answers, referencing past purchases and open support tickets to automatically provide updates and resolve simple questions.

Sales enablement

Many sales technology solutions are already using AI, with natural language processing being the most common. In 2023 and beyond, we’ll see greater adoption of tools using AI to better recommend content and actions and predict the next actions of buyers.

AI can help with the automation of tasks, removing mundane or tedious workloads from your team. This removes time restraints and room for human error. It will also speed up communication inside your organisation.

LXA’s State of Sales Enablement 2022/2023 found that sales leaders were planning to invest in two emerging sales enablement technologies over the next 12 months. 

Tech tools using AI are high on the shopping lists of CSOs, CGOs and CROs, thanks to the potential to automate workflows and improve customer experience. 54% will be investing
in AI-enabled meeting scheduling, which helps salespeople to optimise busy schedules and find gaps for key meetings.

Common uses of AI in sales enablement include:

• Demand forecasting. Forecasts are complex, but they can be automated. AI enables the creation of automated and accurate sales projections based on previous client interactions and historical sales results. Accurate sales forecasting is often made more difficult thanks to a lack of high quality data, over-optimism on the part of salespeople, and unpredictable sales cycles. AI can help to address these issues through iterative learning, more accurate data collection through chatbots, and consistent forecasting methods.
• Lead scoring. AI can help sales teams to prioritise leads based on the probability of conversion. AI can rank the opportunities or leads in the pipeline based on their chances of closing successfully by compiling historical information about a
  client, and the salesperson’s customer interaction history.
• AI-guided sales coaching. AI helps augment human sales coaching. AI software can monitor speech patterns to detect energy levels, tone, and speaking pace to
  offer instant feedback to salespeople. AI in sales training can also streamline the onboarding process and fill those gaps where sales managers and leaders don’t
  have the time or resources. It can be delivered just in time, at the point when sales reps need it to help enhance their product knowledge.
• Sales intelligence. Sales revolves around data, for lead qualification, understanding customers, and assessing the effectiveness of the sales enablement process. Sales intelligence tools help teams to monitor data and identify triggers for sales teams. Sales intelligence can not only save hours of research time but also enables salespeople to improve their preparation for pitches and proposals. AI
  can assist in the automation of many of these processes, improving the quality and accuracy of data for example.
• Greater efficiency. AI-powered tools can help teams avoid repetitive tasks, such as scheduling meetings. It enables reps to spend more time selling and connecting with buyers instead of spending their time just prepping to sell. For example, AI
  meeting scheduling tools make it easier to manage their own calendars and for customers to get in touch as needed, blocking out times when sales reps are unavailable and speeding up the booking process.
• Communication with prospects. It is very important how and when you reach out to your customers. AI can help. AI can enable chatbots to work well with the help of natural language processing and can be available around the clock. Chatbots
  can be useful when a customer needs a quick answer and can buy some time until a sales rep is available. AI extends beyond chatbots. It can also provide salespeople with important info about potential customers that can lead to substantial improvements in communication. The algorithm can figure out the best time to reach out, such as when a website viewer is on the right page. AI can recognise the visitor’s intent, initiate the conversation, and then bring in the sales rep.
• Expert Recommendations. Some AI systems can recommend sales actions, even telling sales teams which actions the system thinks will make the most sense, based on goals and data insights. This might be how to price a deal, whom to target, or which customers to approach with upsells or cross-sells. This takes the time-consuming deliberation out of the sales process, freeing the salespeople up to close deals.

Voice AI

Voice AI tools provide a range of potential uses for marketing and sales, from content
creation to digital commerce.

• AI transcription. Tools like Otter.ai allow easy transcription of conversations. The
  AI element helps to improve the accuracy of transcriptions over time, and the technology allows for the recording and transcription of meeting notes, training sessions, and interviews.
• AI translation. AI translation is becoming more accurate, and with it the potential uses for business increase. It can be used to create marketing campaigns across different territories, or for website content localisation.
• AI voice generation. Voice generators convert text to speech and can streamline the content creation process, by allowing marketers to quickly create audio content for use in videos, presentations and voice ads.
• Voice commerce. With the number of digital voice assistants predicted to reach 8.4 billion by 2024, the market for voice-enabled commerce is growing. Alexa and other voice assistants already enable voice commerce, and brands such as Honda have introduced their own voice commerce apps to purchase fuel and parking services.

Analytics and insight

Tools using AI can now automate and optimise analytics and insight across areas such as competitor tracking, business intelligence, and predictive analytics.

• Predictive analytics. Predictive analytics uses data to forecast future outcomes. The process uses data analysis, machine learning, artificial intelligence, and statistical models to find patterns that might predict future behaviour.
  
  Predictive analytics enables businesses to optimise inventory, improve delivery times, increase sales and ultimately, reduce operational costs. The addition of AI, produces more accurate and timely forecasting that can react to rapidly changing circumstances.
• Business intelligence. AI in business intelligence can help to break volumes of big data into granular insights, with the ability to analyse new data and identify significant trends. Using AI, business intelligence tools can analyse a wider range of data, both structured
  and unstructured, in order to obtain more detailed and comprehensive insights.
• Competitor monitoring. AI-enabled monitoring tools allow businesses to automate competitor tracking processes, such as monitoring social posts, products, prices and reviews at scale.

AI initiative development

With the buzz around AI recently, it’s understandable that marketers and sales teams will look to see how they can use AI technology to improve their own performance.

However, it’s important to take a strategic approach to the use of AI, as with any other investment in marketing or sales technology.

One key point to consider is that AI is already in use across a range of marketing and sales platforms, some of which you may already use in your tech stack.

At LXA, we would recommend the following process to begin AI-related initiatives.

• Consider business goals. What are you looking to achieve through deploying AI? It’s important to be clear about where your business is heading, and how AI can help you to get there. Whether the goal is to improve workplace efficiency or productivity, you should consider the many benefits of AI and how they align with your business goals.
  
  - What are our top business priorities?
  - What problems do we want or need to solve?
  - How can AI help us to deliver our strategic goals?
• Use cases. Consider the use cases for AI. There may be many possible use cases, but it can help to focus on a smaller number to ensure goals are achievable and realistic. These use cases may be:
  
  - Making business processes for sales or marketing more intelligent.
  - Automating repetitive tasks to free up time and resources.
  - Improving customer experience through greater personalisation.
• Audit current tools, capabilities, adoption and utilisation. Audit your existing tech stack for tools which may offer AI-based capabilities. Some estimates suggest that up to 80% of software features are not used, so it’s possible there are features that match your AI use cases. Training may be required to help teams understand new features and drive adoption, and processes may need to adapt. 59% of respondents to a survey by Narrative Science named ‘shortage of data science talent’ as the primary barrier to extracting value from their big data technologies.
• AI initiative alignment and priority. It may take some time to deliver on the AI use cases you’ve identified, so it can help to also identify a few quick wins, prioritising use cases that enable you to demonstrate value and gain buy-in for bigger AI initiatives.
  
  - Marry desired use cases to existing tool sets and focus on enablement, adoption, change management and training to drive use. If AI capabilities are there, then focus on the skills and processes to deliver on them, and drive greater use.
  - Identify tools with AI that offer off-the-shelf capabilities to answer use cases. Tools with AI built in can deliver a quick win.
  - Explore custom AI tool platforms. Where the AI-related capabilities are not available in existing platforms, there are many custom AI tools that provide alternative options.

The future of AI in marketing, sales, CX and operations

It’s no exaggeration to say that AI will have a transformative effect on marketing, sales
and customer experience over the next decade. For example, if Generative AI continues to develop at its current rate, it will have huge implications for content creation, SEO and more.

Google, Microsoft, Amazon, Meta and other tech giants are investing heavily in AI technology, as they see it as central to their future business models.

Microsoft CEO Satya Nadella recently stated that AI would eventually be included in all of the company’s applications. It hopes its moves into AI, including the partnership with
OpenAI, will give the company the edge over rivals.

According to Google CEO Sundar Pichai, ‘AI will have a more profound impact on humanity than fire, electricity and the internet’.

With this rush to build AI applications, it’s important to consider the ethical implications. Google, Microsoft, Meta et al are taking a ‘build fast and ask questions later’ approach with technology that’s likely to spark profound changes in society.

For example, the algorithms powering AI applications are vulnerable to the same biases as the humans that train them. There are other concerns around privacy, copyright and the effect on the workforce, as well as the broader impacts on society.

Gartner predicts that, by 2025, 70% of enterprise CMOs will see accountability for ethical AI in marketing among their biggest concerns.

While marketing and sales professionals need to consider the ethical impacts as part of their wider strategy, the fact is that AI is here right now, and can be used to improve business performance.

It’s not a question of future-gazing, the challenge for marketing and sales is to use the AI technology that exists within tech stacks, and can be used today to create better customer experiences.

Glossary of terms

• AI-powered CX. AI technologies can improve the customer experience through
  more effective personalisation, and better service. The use of intelligent chatbots
  for customer service is one example.
• AI-powered growth. The use of AI to improve overall business growth, through the introduction of more efficient processes, and improved sales and conversions.
• AI-powered marketing. The use of AI technologies to make automated decisions based on data collection, data analysis, and trends to optimise the performance
  of marketing.
• AI-powered sales. Using AI to improve sales performance through automated
  analysis and learning of customer behaviour.
• Artificial general intelligence (AGI). A (for now) theoretical form of AI where a machine would have an intelligence equal to humans. Potentially the next big
  breakthrough in AI technology.
• Big data. Data that is so large or complex that it’s difficult or impossible to process
  using traditional methods.
• Data engineering. The building of systems to enable the collection and usage of data. This data is used to enable subsequent analysis and data science.
• DataOps. A collection of technical practices, workflows, and architectural patterns
  that enable higher data quality, clear measurement and monitoring of results, and rapid experimentation and collaboration.
• Data Science. Data science uses maths and statistics, advanced analytics, AI and machine learning to uncover actionable insights from an organisation’s data.These insights can be used to guide planning and decision making.
• Machine learning (ML). ML is a branch of AI which focuses on the use of data and algorithms to imitate the way that humans learn, gradually improving its accuracy.
  The algorithms used to produce content recommendations for Spotify and Netflix are examples of ML.
• MLOPs. MLOps (machine learning operations) is a set of practices for collaboration and communication between data scientists and operations professionals.
• Reinforcement learning (RL). A subset of machine learning that allows an AI-driven system to learn through trial and error using feedback from its actions. RL
  learns from mistakes and offers AI that mimics natural intelligence as closely as it is currently possible.
  
  

Citations

1 Source: Sarah O’Neill, ‘AI and ML in B2B: Stats and Trends for 2023’, LXA, December,
2022. (https://www.lxahub.com/stories/ai-and-ml-in-b2b-stats-and-trends-for-2023)
2 Source: Layla Nelson, ‘ChatGPT: World’s most powerful AI chatbot will soon “look like
a boring toy,” says OpenAI boss’, Canada Today, December 27, 2022. (https://canadatoday.
news/ca/chatgpt-worlds-most-powerful-ai-chatbot-will-soon-look-like-a-boring-toy-saysopenai-
boss-204101/)
3 Source: BBC, ‘Google AI beats human Go champion’, BBC, May 25, 2017. (https://www.
bbc.co.uk/news/technology-40042581)
4 Source: MetaAI, ‘MetaAI Presents Cicero’. 2022. (https://ai.facebook.com/research/
cicero/)
5 Source: LXA, ‘The State of Martech 2022/2023’, LXA, October 2022. (https://www.
lxahub.com/the-state-of-martech-2022/2023)
6 Source: Dave Rogenmoser, ‘Jasper Announces $125M Series A Funding Round’,
Jasper, November 10, 2022. (https://www.jasper.ai/blog/jasper-announces-125m-seriesa-
funding)
7 Source: Manash Singh, ‘Google acquires Twitter-backed AI avatar startup Alter for $100
million’, TechCrunch, October 27, 2022. (https://techcrunch.com/2022/10/27/googleacquires-
twitter-backed-ai-avatar-startup-alter-for-100-million-source-says/)
8 Source: Pencil ( https://www.trypencil.com/)
9 Source: Abert (https://albert.ai/ai-marketing-software/)
10 Source: Varos (https://www.varos.com/)
11 Source: Drift (https://www.drift.com/about/)
26
12 Source: Kalhan Rosenblatt, ‘ChatGPT passes MBA exam given by a Wharton
professor’, NBC, January 23, 2023. (https://www.nbcnews.com/tech/tech-news/chatgptpasses-
mba-exam-wharton-professor-rcna67036)
13 Source: Sonya Huang, ‘The Generative AI Application Landscape’, Twitter, October 17,
2022. (https://twitter.com/sonyatweetybird/status/1582040028015837187_)
14 Source: Matthew Dixon, Lara Ponomareff, Scott Turner, and Rick DeLisi, ‘Kick-ass
customer service’, HBR, January 2017. (https://hbr.org/2017/01/kick-ass-customerservice)
15 Source: LXA, ‘State of Sales Enablement: Sales Operations and Salestech
Report 2022/23’, LXA, November 2022. (https://www.lxahub.com/state-of-salesenablement-
2022/23)
16 Source: WRAL Techwire, ‘Pendo study: With 80% of features not used, software
execs re-evaluating success’, WRAL Techwire, January 28, 2020. (https://wraltechwire.
com/2020/01/28/pendo-study-with-80-of-features-not-used-software-execs-reevaluating-
success-metrics/)