Prepare for Your First Jump Into AI for Marketing and Sales by Minimizing Risk

Included Research:

AI Used in Marketing, Sales, and Customer Success? Is It New?

Consumers and businesses have been employing AI and machine learning for decades. From Amazon’s home page and Facebook’s natural language search to Google Photos, Gmail’s Smart Compose, and Waze’s AI-based traffic predictions and routing, consumers have routinely been exposed to AI. Big Data and expensive computing power made AI only accessible to large companies. This has changed with broader access to LLMs that, like ChatGPT, can be used via browser.

The most frequent business use case has been in data analytics and advertising. Marketers today can go to Google Analytics and ask for data in very natural language. The AI elements of Google Analytics also present them with some predictive reports and queries. Google Ads uses AI to help marketers write ad copy headlines by providing suggestions and prompts. Behind the scenes, the Google advertising platform’s AI collects massive amounts of data to predict buyer intent and present consumers with the most relevant ads.

Software as a service (SaaS) companies have also been exposed to AI in their software platforms from customer data management. IBM Watson and Salesforce Einstein are a few other examples. A 2020 Deloitte global survey of early AI adopters showed that three of the top five AI objectives were oriented toward marketing: enhancing existing products and services, creating new products and services, and enhancing customer relationships.

B2B Companies Using AI in Marketing, Sales, and Customer Success Will Create Faster and More Powerful Scaling Opportunities

Recommended strategy for considering AI in marketing, sales, and customer success

Implementing the right AI solutions will deliver significant benefits to your business. The first thing to do is make a business case to help you prioritize and justify the investment. Make sure your business case falls into these buckets.

1. Revenue: AI can help personalize marketing and sales interactions at scale, leading to higher conversion rates and customer lifetime value. Implementing AI chatbots, recommendation engines, and predictive analytics can increase revenue and flatten staffing costs.
2. Efficiency: AI can automate high-volume, repetitive tasks to enable employees to focus on higher-value work. Intelligent process automation with AI reduces human errors and saves time.
3. Automation: By automating routine tasks with AI, we can reduce mundane work and boost employee satisfaction and retention. Employees can spend time on more strategic initiatives.
4. Cost: AI adoption can save costs through increased workforce productivity and automation. Fewer human hours are needed for routine tasks, resulting in lower costs. AI also helps detect fraud and minimize risk, further reducing potential costs.

These use cases address existing problems. Here are a few examples of how they have been implemented.

Sales and marketing frequent use cases:

• Reducing the sales cycle by predicting the leads with the highest intent and increasing revenue.
• Creating a cohort of existing customers and creating ideal customer profiles (ICP). By continuous learning, predict changes in ICP or customer segments.
• Creating briefing books for sales visits to enterprise customers, eliminating the need to spend hours of manual work.
• Creating meeting minutes and action items for sales and demo calls and saving time for the sales team to increase their capacity.
• Customize and create links to sales enablement content based on the sales interactions with the prospect, helping increase conversions.
• AI-enabled audience segmentation for marketing campaigns reduces work in manually updating lists and increases targeting accuracy.
• Chatbots for lead development, customer support, and cross-selling or upselling.
• Inbound call analysis and routing, customer comment and email analysis, classification, and response.
• Marketing campaign automation (including emails, landing page generation, and customer segmentation).
• Marketing mix analysis.
• Online product merchandising.
• Pricing optimization.
• Product or service recommendations and highly personalized offers.
• Programmatic digital ad buying.
• Sales lead scoring.
• Social-media planning, buying, and execution.
• Social-media sentiment analysis.
• Television ad placement (partial).
• Web analytics narrative generation.
• Website operation and optimization (including testing).

Mini Use Case Examples From the Journal of Business & Industrial Marketing

Customer Knowledge: Inventory of activities for creating, codifying, sharing, and applying knowledge about customers, such as the what, how, and why of the purchasing decision, and the antecedents and consequences of this purchasing decision.

• Profiles: Creating a comprehensive profile of current or potential customers. AI can use structured and unstructured data inputs of several types, such as recency, size, frequency, the type of past purchases, current web browsing behavior, psychographic and demographic characteristics, and interactions with the firm to create this profile. Using machine learning and predictive algorithms, the resulting profiles of current or potential customers can then be applied to improve customer relationship efforts and for prospecting future customers.
• Using predictive models in the sales funnel: AI systems can be used for prospect scoring to evaluate potential customers based on their likeliness to purchase and to identify high-quality leads. This is a task that requires a considerable number of human resources. AI can also automate some more routine tasks during the pre-approach and approach stages, such as scheduling meetings or answering frequent questions using chatbots. In the presentation and close stages of the sales funnel, AI-presentation bots can help sales staff create compelling presentations. Additionally, AI can help overcome customer objections by using emotional AI to understand client responses at this stage or by using AI-enabled battle cards to undermine competitors and strengthen the firm’s value proposition.

User Knowledge: Using AI to create and codify insights about users’ experiences, product usage, user attitudes, values, wants, and future needs.

• Product development: Using AI to create a psychographic profile by identifying sentiment, emotions, values, and expressions provided by the users via text, voice, email, and social media. This would be a valuable source of insight for B2B marketers for innovation and new product development.
• Product usage: Identify themes and patterns of product usage and how users creatively alter products to facilitate new product development.

Market Knowledge: Using AI systems with natural language processing and machine learning algorithms to get external market data by analyzing online content, social media, blogs, press releases, etc.

• Competitive intelligence: Identify keywords and themes from public data sources such as competitors’ press releases and social media posts, and use these insights to create battle cards for sales and marketing positioning documents.
• Fake news alerts: Creation and dissemination of fake news can threaten the brand and reputation. Using AI insights, marketers can be vigilant about how their brand is associated with fake news and develop effective tactics to manage these threats.

What technical resources are needed to implement AI in marketing, sales, and customer success?

Recent advances in AI, machine learning, and natural language processing have made these technologies more accessible to businesses of all sizes, thanks to a growing number of providers and resources. Companies looking to leverage AI have four primary options:

1. Leveraging internal data scientists and tech resources to set up, maintain, and run AI models for the stakeholders in sales, marketing, or customer success. This involves internal costs like payroll, technology expenses, and recruitment.
2. Choosing a software platform with AI incorporated while choosing your tech stack.
3. Integrating an AI module into existing marketing software to gain valuable insights and take data-driven actions to enhance marketing strategies and achieve better results.
4. Using a managed AI service provider that develops, builds, and maintains custom AI models for each use case. The models have a setup fee and an ongoing learning and maintenance fee. The costs usually depend on volume and can vary by use case.

Companies must weigh factors like use case complexity, data infrastructure, and budget when choosing the best AI approach for their sales, marketing, and customer success needs. Each approach has trade-offs – building in-house allows customization but requires significant internal resources, while managed services offer turnkey AI solutions but with recurring fees based on data volumes.

Get Started With Artificial Intelligence | Info-Tech Research Group (infotech.com)

Reduce Fear of AI Among Employees

Technology has always had an impact on jobs and education. For example, the Pony Express no longer exists due to the invention of the automobile. Similarly, in most business meetings the role of a human transcriber/stenographer has largely been replaced by voice-to-text technology.

AI can make employees happier by automating tasks that are laborious and time-consuming. To identify such tasks in your team, consider the following:

• AI can automate manual and repetitive tasks, allowing employees to focus on more strategic work. For example, one of your team members may be responsible for downloading prospect lists from three software programs, combining them, and segmenting them by customer engagement. This is a manual and repetitive task that AI can easily automate. AI can also be used to identify unsubscribed and churned customers, saving your team time, and improving your data’s accuracy.
• AI can also be used to generate insights and predictions from data. For example, AI can predict which customers are most likely to churn or which marketing campaigns are most likely to be successful. This information can be used to make better decisions about your marketing and sales efforts.

Manually updating and analyzing enormous amounts of data is a tedious and error-prone task that can lead to employee dissatisfaction.

Some examples of how AI-powered automation can be used to improve the workplace:

• Analyze substantial amounts of data and identify trends and patterns to lessen employee efforts to do this manually.
• Automate data entry to save employees a significant amount of time and effort and reduce errors.
• Improve accuracy and quickly summarize insights by generating reports automatically.

AI in Marketing: How to Avoid the Pitfalls and Seize the Opportunities

Dangers of using publicly available AI models:

• There are several dangers to using publicly available AI models. One of the first steps leaders should take is to familiarize themselves with this technology and train and educate their teams on the right way to use AI.
• Use of AI that is free and open to all and trained on public data has legal implications and does not give a competitive advantage. Without proper guidelines, the pitfalls of inadvertent sharing of customer data or internal proprietary information can lead to:
  • Customer data breaches
  • Regulatory fines
  • Loss of customer trust
  • Legal liability and financial damages

Examples of risk:

• One of your employees wants to test the capabilities of ChatGPT and wants to learn how to use it. The employee downloads a small customer list in Excel and uploads it to ChatGPT.
• They would have inadvertently violated data privacy regulations that could subject the company to a heavy fine.
• The number one concern in posting any proprietary information into open public LLMs is that anyone using similar prompts can access that information, including your competitors and bad actors.
• Publicly available LLMs have been trained on publicly available information without the ability to ascertain facts from fiction. They come with disclaimers that human analysts may look at your information.

Case study: A top electronics company banned the use of ChatGPT after three issues were discovered. These issues were:

1. Using ChatGPT to help fix problems with their source code. With the right prompts, anyone could discover this code, and the code is also now the property of OpenAI.
2. Uploading internal meeting notes relating to their hardware to convert meeting notes into a presentation, thereby exposing proprietary data.
3. Asking ChatGPT to optimize test sequences for identifying faults in chips, which could help other competitors without having to do their own research.

Publicly available LLMs like ChatGPT have limited use cases and can be risky to rely on. Although they might seem harmless for simple searches and opinions, their tendency for bias, misinformation, and lack of accountability makes fact-checking essential. We have not seen any use cases for benefits from using openly available LLMs.

For businesses seeking greater control and confidence, investing in a custom AI model trained on your own data offers a safer path. Although data size can initially be a challenge, advancements in data augmentation and transfer learning are rapidly paving the way for smaller companies to reap the benefits of personalized AI.

How to Become an AI-Powered Sales, Marketing, and Customer Success Organization

While AI has risks, it is also a wonderful opportunity to show your innovative thinking mixed with adequate planning and caution. Inaction is not an option. Jumping in with a planned and structured approach is safe.

The Info-Tech approach:

• Give AI in your organization a purpose aligned to business value.
• Develop an iterative, structured, and reusable process to identify, rationalize, and prioritize potential AI use cases.
• Start with a pilot before expanding out to new processes and lines of business.

Following this note will be a series of notes and a comprehensive blueprint that will walk you through using AI with successful use cases in several aspects of sales, marketing, and customer success.

See the related resources below for resources to share with your CIO, COO, CTO, and other business leaders who will be critical to the success of AI in marketing, sales, and customer success.

An accompanying AI evaluation tool will be helpful to the CEO and functional leaders in early-stage, mid-stage, and at-scale companies to evaluate shortlist and prioritize the AI in marketing, sales, and customer success projects.

Related Resources for Use of AI in Sales, Marketing and Customer Success

Artificial Intelligence Research Center

The Era of Generative AI C‑Suite Presentation

Get Started With Artificial Intelligence

AI Trends Report

Build Your Generative AI Roadmap

Navigating the Frontier: Generative AI’s Promise and Peril for Product and Product Marketing Managers

SoftwareReviews has a list of AI/ML vendors at Best Machine Learning Platforms | SoftwareReviews, and you could talk to an analyst to learn more. Other lists include The AI 50 2022 | Forbes.com and Top AI as a Service Companies | Venture Radar.

Works Cited:

Paschen, Jeannette, et al. “Artificial Intelligence (AI) and Its Implications for Market Knowledge in B2B Marketing.” The Journal of Business & Industrial Marketing, vol. 34, no. 7, 2019, pp. 1410–19, https://doi.org/10.1108/JBIM-10-2018-0295.