An unexpected pattern emerged from our recent survey of B2B tech professionals regarding their thoughts and experience using AI in their roles.

We asked a group of sellers with different levels of seniority, sales enablement professionals, marketers, product managers and developers, and customer success managers six questions:

  • What is your job title?
  • Have you used or experimented with AI tools at work yet?
  • If so, how do you use it?
  • Which applications of AI are most helpful?
  • Which are least helpful or when to avoid AI entirely?
  • How does AI improve or detract from the buying experience?

While 65% of the total group responded “yes” to using AI, the percentages of those responding in the affirmative either increased or decreased according to role. Here’s the final breakdown:

  • HR: 100%
  • Marketing and creative: 71%
  • Product developers and managers: 70%
  • Customer Success: 67%
  • Sales (including enablement): 65%

Notice a pattern here? We did. 

It appears that the closer someone is to the customer, the more wary they are of using AI — with sales and customer success being the most reluctant groups to implement generative AI in their daily work.

Why? A closer look at the responses to the final question “How does AI improve or detract from the buying experience?” showed concerns about losing rapport with customers due to less personalized email responses, a general lack of authenticity and the “human touch,” and issues with information accuracy. 

This data led us to ask: what is the impact of AI on the buying experience? And how can we make sure it’s a positive one at the individual and institutional levels?

AI’s impact on the buying experience

Let’s look at a typical buying journey for a SaaS product and how AI could impact each step and the people responsible for that stage of the purchase cycle (journey source).

Step  Customer action Primary role impacted Type of AI AI’s application & impact
1 Need recognition/


Marketing -Generative AI

-AI predictive and recommendation engines

AI can analyze vast amounts of customer data to identify patterns and trends. Marketers can use this information to make data-driven decisions and predictions about customer behavior. For example, AI can predict which customers are most likely to make a purchase, which products are likely to be popular, and when customers are most likely to buy. 

 This enables marketers to optimize their marketing and content strategies and allocate resources effectively to attract the right customers for the following stage — “information gathering,” so the right customer is matched with the right solution from the outset, reducing effort and frustration for the customer.

 Marketers can also deploy generative AI tools to help with research, brainstorming, and draft writing.

2 Information search Marketing -Generative AI

-AI recommendations


By creating personalized and targeted content (with the help of generative AI when needed), marketers can optimize their output so their target buyers are served their content when engaging with search engines and chatbots.

Using AI can cut down on research time for customers.

3 Evaluation of solutions and alternatives Marketing and sales  -Generative AI

-AI recommendation engines


-Conversational AI

-Coaching AI  

AI-powered recommendation engines can analyze customer data and provide personalized product recommendations. 

These recommendations can be based on a customer's browsing history, purchase history, and preferences. By suggesting relevant products or services, AI can enhance the customer's buying journey and increase the likelihood of a purchase.

4 Appraisal Sales -Generative AI

-AI recommendation engines


-Conversational AI

-Coaching AI  

During the appraisal stage, customers will be regularly engaging with sales reps, asking questions, and assessing product/need fit. 

Sellers can prepare for and improve customer interactions during this stage by practicing with and getting feedback from conversational and coaching AI that objectively evaluates their speaking patterns, information provided, response times, emotional output, and more, all of which translates to a more polished seller for the buyer to interact with. 

AI can also summarize meetings, automatically log data to the CRM, and highlight the questions customers asked to show sellers what content or info to follow up with.

They can also use generative AI to quickly summarize and digest content, draft emails, and see recommendations for what content to share with the buyer.

Customers may also employ chatbots during this stage. Unlike human agents who have limited working hours, chatbots can be available 24/7. This means that customers can access information and support at any time, even outside of regular business hours. Chatbots ensure that customers can gather information whenever they need it, enhancing their overall experience.

5 Purchase Sales -AI recommendation and predictive analytics


AI can help salespeople identify potential customers, track their interactions, and predict when they will likely make a purchase. By focusing their efforts on the most promising leads, salespeople can close deals more quickly and efficiently. 

Chatbots can also help speed up sales cycles by helping customers get quick answers to questions without waiting for an email or meeting with a sales rep, or even letting sellers know the best time and channel to reach a customer based on past behavior

6 Post-


Customer success and support -Generative AI

-AI recommendations


AI can assist agents by providing them with relevant information and insights to help them resolve customer issues more quickly and efficiently. In this way, it can help improve onboarding, implementation, and adoption. 

“There are many gen AI use cases after the customer signs on the dotted line, including onboarding and retention. When a new customer joins, gen AI can provide a warm welcome with personalized training content, highlighting relevant best practices. A chatbot functionality can provide immediate answers to customer questions and enhance training materials for future customers.” -McKinsey

7 Renewal/


Customer success and sales -AI recommendation and prediction engines


Customer success and account managers can use AI to analyze customer data and provide insights into customer behavior, which can help them understand their customers better and close more sales with existing customers. 

By offering personalized advice and solutions, reps can engage customers and build stronger relationships over time to increase retention and expansion opportunities. 

Customers may also engage AI chatbots to find out about additional solutions from their current provider to expand their usage, alerting their CSMs and AMs to their interest.


To better visualize this, consider the following figure in which the author (Schrotenboer) illustrates the application of two particular types of AI — recommendation engines and chatbots — throughout the buyer’s journey:

AI impact on the buying journey map

How to use AI while ensuring a positive buying experience

Strike a balance between AI and human agents

The overwhelming response from the survey was that AI is helpful to humans in customer-facing roles, but only when used thoughtfully and in a limited capacity that does not degrade authentic human relationships.

One example of balance is to let chatbots field easy questions with “stock” answers, and ping a human to jump in when things become more complex. This is already common practice for many companies who offer 24/7 customer service, like power and internet providers. 

Another example is for humans to always edit AI output, especially when it comes to creating marketing content or sales communications. 

To achieve this balance, you need to take your customers’ preferences into account.

Acknowledge (and ask) for buyer preference

How many times have you hit “0” on the dial pad to speak to a human instead of going through the automated phone system? Sometimes you just need to talk to someone — because your problem or question is too complex or nuanced for a bot to understand, it doesn’t fit into the predetermined options, or because the discussion requires empathy. 

Then there are days where you’d rather just click “2” and let a computer solve your problem, or head for the self-checkout line because you have neither the need or desire to socialize.

Companies implementing AI interactions during the buying journey should offer an “operator” option wherever AI is in place, such as website chatbots, in order to serve every customer according to their preference at any given time.

Protect privacy and sensitive information 

AI-powered personalization relies on collecting and analyzing customer data, which can raise privacy concerns. 

Customers may be uncomfortable with the amount of data that's being collected and how it's being used. When it comes to your customers, stick to proprietary tools that are under your control and have a privacy policy or consent form for customers to review before deploying AI. Let customers who are nervous or work with highly classified information opt out.

AI systems are also vulnerable to hacking and other cyber attacks. Ecommerce businesses that use AI must ensure that their systems are equipped with robust encryption credentials and that customer data is protected.

Be wary of bias and actively work against it

AI algorithms can be biased and discriminatory, particularly if they are trained on biased data. This can lead to unfair treatment of certain groups of customers.

Your internal teams should be in charge of training your proprietary AI behind the scenes and use a process called debiasing (source). This can help ensure that AI is aligned with the company's values and that it isn’t biased or discriminatory.

A possibility in feature-based models is to manually intervene and leave out the features that could lead to biased predictions for a task, e.g. the New York Police Department (NYPD) refrains from using the race of a person to predict the risks of future crimes.

-Towards Socially Responsible AI: Cognitive Bias-Aware Multi-Objective Learning by Procheta Sen and Debasis Ganguly

Your HR, learning and development, or enablement team can also provide bias awareness training when introducing AI into your operations.

In this figure, McKinsey offers “Six potential ways forward for AI practitioners and business and policy leaders to consider [regarding bias]:”

McKinsey AI Bias chart

Ensure data accuracy 

When commercial leaders were asked about the greatest barriers limiting their organization’s adoption of AI technologies, internal and external risk were at the top of the list. From IP infringement to data privacy and security, there are a number of issues that require thoughtful mitigation strategies and governance. The need for human oversight and accountability is clear, and may require the creation of new roles and capabilities to fully capitalize on opportunities ahead.

- McKinsey: AI-powered marketing and sales reach new heights with generative AI

Make sure AI is not providing, either directly or through an agent seeking answers to customers questions, inaccurate information that leads a customer to come to one conclusion that is disputed by a human rep later on, leading to conflict. 

This is equivalent to overpromising and under-delivering, causing angry and disappointed customers to churn down the line, wasting everyone’s time and money. 

Similar to debiasing, you will also want staff to train any AI chatbots or other tools you make available to customers to offer only accurate, up-to-date information that is in line with the messaging your human reps use. 

Key takeaways for managing AI impact 

It is not only important for companies to apply technological advancements like artificial intelligence; perhaps it is even more important to thoroughly understand those techniques and their impacts in order to apply it with the utmost precision and accuracy.

- The Impact of Artificial Intelligence along the Customer Journey

For the individual: Start experimenting with AI now to get a feel for when it helps you and your customers and when it doesn’t. This way you are prepared when AI use becomes commonplace and you already know when — and when not — to use AI throughout your day and throughout the customer journey.

For the institution (and the people leading them): Have an AI strategy and policy for your organization, especially for customer-facing roles. Treat AI usage as a skill and offer employee training in areas like prompt writing, AI response editing, fact-checking, and bias awareness. 

If you want the productivity gains it promises, start experimenting and getting educated about AI now. Because soon all of the software in your company’s tech stack will incorporate AI in some form, and customers will expect it in theirs.