The AI Teammate: How to Collaborate With Artificial Intelligence in Sales

The AI Teammate: How to Collaborate With Artificial Intelligence in Sales

A new player has entered the field of B2B sales: Artificial Intelligence (AI), specifically in the form of large language models (LLMs).

These advanced AI systems, which power a variety of chatbots and AI assistants, are revolutionizing the way salespeople work. But far from replacing human sellers, these AI tools are emerging as powerful collaborators—teammates that can enhance your capabilities and help you achieve new levels of productivity and success.

Here are some tips on how sales professionals leverage LLM-powered AI as collaborative partners and how you can do the same—using AI assistants such as ChatGPT, Claude, and Bing (Copilot), and sales-specific tools, such as Salesforce Einstein and Microsoft's Copilot for Sales.

The Mindset Shift: AI as Your New Teammate

RAIN Group's AI in the Sales Process research revealed an important insight: sales professionals who view AI as a collaborative teammate rather than just a tool are seeing a significantly greater impact on their performance and productivity.

Those reporting the most significant impact from AI are 4.8 times more likely to strongly agree that AI acts as a new teammate to collaborate with.

That mindset shift is key to unlocking the potential of AI in your sales process.

Think of AI assistants as the ultimate support staff—always available, incredibly fast, and capable of processing vast amounts of information. It's like having a personal assistant, researcher, and analyst all rolled into one, working tirelessly to support your sales efforts.

How to Collaborate With Your AI Teammate

1. Brainstorming and Ideation

Use AI to generate fresh ideas and perspectives. Whether preparing for a client meeting or developing an outreach campaign, AI can help you think creatively.

Tip: Try prompting your AI assistant with questions like, "How can I use AI as a teammate to collaborate with in my prospecting efforts?" or "Generate a list of potential objections and creative responses for our new product launch."

2. Research and Information Gathering

Use AI to quickly gather and synthesize information about prospects, industries, and market trends. Doing so can help you enter every conversation well-informed and prepared. Our research finds that sellers using AI are 3.4 times more likely to strongly agree that it helps uncover challenges and pain points.

Tip: Ask your AI teammate: "Summarize the latest industry trends affecting [prospect's company]" or "Compile a brief on [prospect's company]'s recent financial performance and strategic initiatives."

You can use various prompts to give chatbots that can assist in gathering information.

3. Content Creation and Customization

Collaborate with AI to draft and refine your sales communications. AI can help you create more engaging and personalized content, from emails to proposals. In fact, 68% of respondents in our study strongly agree or agree that AI improves value messaging in client interactions.

Tip: Use AI to generate outlines or to suggest improvements to your email drafts. For example, "Review this email draft and suggest ways to make it more compelling and personalized for [prospect]."

4. Meeting Preparation and Follow-up

Work with your AI teammate to prepare for meetings and process the outcomes. AI can help you develop question lists, create agendas, and summarize action items.

Tip: After a meeting, ask your AI assistant: "Generate a summary of key points discussed and a list of follow-up actions based on these meeting notes."

5. Problem-Solving and Solution Recommendations

Use AI as a sounding board when tackling complex sales challenges. It can help you analyze situations from multiple angles and suggest potential solutions. Our research finds that 60% of respondents strongly agree or agree that AI provides valuable sales insights.

Tip: Present a challenging sales scenario to your AI teammate and ask, "What are three different approaches we could take to resolve this situation?"

Overcoming Challenges in AI Collaboration

Although the benefits of collaborating with AI are clear, it has its challenges. Here are some common hurdles and how to overcome them:

• Accuracy concerns. Fully 59% of respondents in our study expressed concern about inaccurate or misleading information from AI. Always verify information provided by AI. Use it as a starting point, not the final word.
• Maintaining authenticity. AI is a tool to enhance your abilities, not replace your unique human touch. Use AI-generated content as a foundation, then personalize it with your insights and experience.
• Overreliance. Find a balance between relying on AI and maintaining your critical thinking skills. Use AI to support your decision-making, not to make decisions for you.
• Learning curve. Half of respondents reported difficulty keeping up with AI advancements. However, those reporting the greatest impact are 3 times more likely to use AI tools daily or frequently. By committing to regular use, you'll be better positioned to keep pace with AI as it evolves. Take the time to learn how to prompt and interact with your AI tools effectively; the better you become at "communicating" with AI, the more valuable it becomes as a teammate.

The Future of AI Collaboration in Sales

As AI technologies advance, the potential for collaboration will only grow. By embracing AI as a collaborative teammate today, you're not just improving your current performance—you're future-proofing your career in sales.

Research shows that those reporting the most significant impact from AI are 3.3 times more likely to strongly agree that it reduces time spent on manual tasks and 3.1 times more likely to strongly agree that it allows focus on other higher-value tasks.

The most successful salespeople of tomorrow will be those who learn to collaborate effectively with AI today. By shifting your mindset to view AI as a teammate, you can unlock new productivity levels, creativity, and success in your sales career.

One-fifth (20%) of sales teams are already using AI tools frequently or daily; now is the time for you to join the ranks of AI-empowered sales professionals.

-----------------------------------------------------------------------------

If you need help with your email, web site, video, or other presentation to promote your company, product, or service, please give me a call at 330-815-1803 or email me at john@x2media.us

Until next month. . . .remember. "you don't get a 2nd chance to make a 1st impression." Always make it a good one!!

Forget AI Sentience, Authenticity Should Be Our Focus in Marketing

Forget AI Sentience, Authenticity Should Be Our Focus in Marketing

As AI penetrates deeper into both business and personal life, it's hard not to be impressed with its fluidity, knowledge, and (usually) natural conversation.

So it's been a popular question: Is AI sentient?

AI programs are often given friendly names like Alice, Leo, and Celia to subconsciously endear them to us, and even lifelike romantic chatbots have even helped many find companionship in lonely times. Does it really matter whether it's artificial or not if it makes you feel a little less alone?

And the Turing test isn't even reliable anymore; ChatGPT has recently passed it. And in a recent study, 54% of participants couldn't distinguish AI chatbots from real people.

Honestly, the "big question" of whether AI is—or could become—sentient is not en mode anymore. What is more pertinent now is what voice we should allow it to have.

We all have carefully built up talented teams and want their trusted voices to still shine through in the work we do. Authenticity is something we're not willing to compromise, and AI is something we can't afford to ignore. So where do we draw the line?

That is the biggest AI question, not its sentience.

The Search for Realness

What does it even mean to be authentic? We hear the word thrown around a lot by people, from looking for authentic partners on dating apps to seeking out brands that speak authentically to them.

Authenticity, in many ways, means to be human. So, if we're seeking to market with authenticity, then it would follow that we need humans behind the work to drive this somewhat illusive feeling of relatable real-ness.

But remember what was said about AI earlier? Sometimes it doesn't matter if it technically qualifies as being sentient or not. If a lonely person can find an authentic connection with a bot, then a customer could find an authentic connection with AI content. If it seems real, if its results can impact us in the same way, the technicalities become moot.

Well, to some degree. Authenticity and AI aren't mutually exclusive. The master marketers pulling the strings behind the curtain, need to recognize the power of the tools that are used while being careful not to slide into uncanny-valley territory.

Prioritizing Human Connection...

With lots of trial and error, you can find out which boundaries work for our individual teams. You can expect to experiment a lot, fail frequently, and succeed occasionally. Some tasks simply need the human touch—and employees behind it with the unique skills to do right by the goals you are pursuing.

Marketers have mixed feelings about AI, so it's imperative to find the right path to build confidence. Here are four areas that need to be mastered by every pro:

1.     Experimentation. One of the most beautiful things about marketing is the opportunity to try new things. Or, rather, the requirement to do so. To captivate customers, you have to think outside the box: Taking risks, working outside the parameters, and breaking expectations should be left to the team.

2.     Customer interaction. Creating authentic connections with clients and customers is the heart of marketing, so this one shouldn't come as a surprise. Building valuable long-term relationships can't be left to the inconsistencies of a chatbot.

3.     Humor and storytelling. Humor is not a specific task, but it's key to creative work. The fun of crafting unique dialogs can be an important part of nurturing the creative mindset. And, of course, storytelling gives life to your work. It comes from the heart, and it's pretty sure AI's isn't fully built yet.

4.     Collaboration. Most business practices rely more and more on collaboration, and it can sometimes be a fragile thing. It can be easy to step on toes or misinterpret actions, so top-tier communication skills are key to taking collaboration to the next level.

...While Embracing the Artificial

Over the last few years, people have found some areas that particularly benefit from its partnership. It seems like everyone has jumped on the train of generic GenAI tools like ChatGPT and Claude, and there's no reason to reject them right off the bat. The truth stands: They save time and they're still novel enough to be fun to play around with.

You still need to make sure to edit and adjust the results, and only using results as references. Properly used, AI tools can be great for brainstorming, refining ideas, or refreshing content, and cross-referencing results with other tools can add another level of quality. Injecting more content and more ideas can bolster creativity and amplify unique voices, rather than speaking over them.

Specialized AI can also be a game-changer for marketers. Generic, large-language models can only go so far. Investing in a tool that can automate tedious tasks can help your team focus on what's important.

Looking Beyond the LLM horizon

Consider options beyond large language models (LLMs). What specialized AI can best support your team?

Depending on your specific goals and challenges, key specialized AI can now execute tasks, make decisions, and showcase creativity; they can be great problem solvers, such as NLP AI for customer sentiment analysis (IBM Watson Tone Analyzer, Brandwatch); predictive analytics AI for forecasting and optimization (Salesforce Einstein, GA4); Computer Vision AI for visual content analysis (Clarifai); recommendation systems for personalization (Dynamic Yield), attribution modeling AI for campaign effectiveness (AppsFlyer); and the most recent: AI agents that, when supplied with high-order marketing goals, can figure out how to accomplish them.

TL;DR: The Illusive Balance

Decide what authenticity means for you. Is it strictly human? Is it only results-based? Your answer will guide your implementation decisions.

Experiment and decide for yourself where you want to draw the boundary. Maybe you're OK with using it more expansively, and maybe you want it to only run in the background. Talk to your team members who are in the trenches to find out where they're not willing to compromise.

AI may or may not be sentient, but you are, and you decide what real-ness looks like. And don't ask ChatGPT for the answer to that one.

-----------------------------------------------------------------------------

If you need help with your email, web site, video, or other presentation to promote your company, product, or service, please give me a call at 440-519-1500 or email me at john@x2media.us

Until next month. . . .remember. "you don't get a 2nd chance to make a 1st impression." Always make it a good one!!

Surveys, Polls, Forms, and Progressive Profiling: How to Write Questions That Deliver Valuable Insights

Surveys, Polls, Forms, and Progressive Profiling: How to Write Questions That Deliver Valuable Insights

As third-party cookies come to an end, brands are focusing on collecting more zero-party data directly from their customers and prospects so they can better understand their needs and wants.

Although larger surveys will play a role in that process, most brands will gain insights a few answers at a time through signup forms, profile pages, polls, and other forms of progressive profiling.

It sounds simple, but writing a good question and collecting reliable answers is harder than it seems. Things can go wrong...

1. Before you write a question

1. When writing the question

1. When writing answer choices

1. Around the timing

1. When analyzing the responses

1. When repeating data collection

• Less than 20%

• 20% to 50%

• More than 50%

• Not sure

Let's talk about best-practices and things to look out for during each of those six steps.

1. Before You Write a Question

It's possible to go entirely wrong before you even begin with your progressive profiling efforts. Consider the following two things.

Always start by understanding exactly what you want to learn from your audience

What's your objective? Why do you want a certain piece of information from them? How are you going to use or operationalize that data point? Does that data give you the insights you want?

That last question gets at a disconnect that many brands struggle with. The goal isn't to collect data. Not really. It's to gain insights you can use to drive the desired outcome. Yes, you need data to get insights, but the two aren't the same thing.

A favorite example of data vs. insight is from B2C marketers who ask customers about their gender. Most of the time that data is used to personalize message content about products for men or women. However, a person's gender doesn't tell you what kind of products they want to buy, because they could be buying primarily for someone else or they could be interested in products for the opposite gender or both genders.

So, the better question is the more direct one: Are you interested in products for men, women, or both?

Don't ask questions you won't act on

Simply asking a question sets an expectation that you'll use that information to make the customer experience better in some way—even if you're just sharing it out with the community.

If you don't do anything with it, however, that can lead to disappointment. And it can lead to lower response rates for future progressive profiling efforts.

Also, gone are the days when you'd collect information because you might need it in the future... at some point... maybe. You don't want the liability of retaining data you're not using, so don't collect it in the first place.

2. When Writing the Question

Once you're clear on your objective, then it's time to craft your questions. Here are some things to keep in mind.

Craft questions that are universally understood

To the degree that it's appropriate, avoid jargon or technical language. If it's needed, consider providing quick definitions in parentheticals.

If your audience is international, think about non-native speakers, who might struggle to understand some long words, colloquialisms, and cultural references.

And finally, use unambiguous time windows, such as saying the past 12 months instead of the past year, which some might interpret as the previous calendar year.

Provide any needed context before the question

The primary concern here is that some people, once they've read a question, will skip to the answer choices because they're in a hurry. (And everyone's in a hurry.)

Another reason is that in the absence of immediate context, people bring their own context to questions, which forces your post-question context to work harder to override the respondent's initial thinking.

Avoid context and introductory statements that might impose a bias on answers

For example, you shouldn't ask, Given the current state of the economy, do you think now is a good time to change supply chain management software providers? You'll get more accurate answers without that introductory clause.

Ask judgment-free questions

Marketers are great at asking leading questions in marketing copy, but you don't want to do that in polls and surveys if you want meaningful results.

Sometimes that means you need an introductory statement or clause that gives the respondent cover to answer truthfully about something that might otherwise make them look or feel bad.

For example, you might preface a question with a clause like Recognizing that you don't have full control over your program... to make it easier for respondents to answer truthfully.

Recognize that people are bad at remembering past behavior

People provide the most reliable answers about now and the recent past. When you're asking about the actions of their organization, things can get even hazier, because the respondent may be relatively new to their company. Consider asking about actions or behaviors from the previous 12 months, at most.

Avoid redundant questions

The more questions you ask, the lower your completion rate will be. So try to ask as few questions as necessary. For example, I saw a recent B2B lead-gen form that asked for both the person's country and world region. If you get the person's country, you can figure out the region of the world, so that question was completely unnecessary.

3. When Writing the Answer Choices

Most likely, the vast majority of the polling or surveying you'll be doing will involve answer choices rather than open-ended questions. So consider the following when crafting those answer choices.

Make answering easy

While this has a lot to do with the questions you ask, the answer choices you provide also have a major impact on how easy a question is to answer.

For example, here's a recurring question I've asked marketers:

What percentage of your company's email marketing revenue is generated by automated and transactional emails?

Ranges make answering the question much easier, because the chances of knowing the exact percentage is low—and you absolutely don't want people to go hunting for information, because they probably won't come back.

Five- and three-point rating scales (e.g., Always, Sometimes, Rarely) generally produce the best results while keeping things easy.

Be careful when using subjective measures

Sometimes, beauty is indeed in the eye of the beholder. Other times, it's not. For instance, some people think that an email deliverability rate of 50% is good, but it's actually horrible.

So, if you asked about brand' email deliverability, you'd likely get very different distributions if you ask whether their inbox placement was Excellent, Good, or Poor versus Over 95%, 90%-95%, or Below 90%.

Provide an N/A option

Even if answering a question is optional, give people the option to not answer the question by selecting N/A, Not sure, or Don't know—or a combination of those, such as Not sure or don't know. Otherwise, they'll guess or put down the answer they think you want to hear, degrading the accuracy of your responses.

4. The Timing

When you ask your audience questions depends on several factors, but the most consequential is whether the responses are useful long-term or short-term.

Answers That Are Useful Long-Term—for Many Months to Years

These include demographic information, such as a prospect's company name or industry, and technographic information, such as details about their tech stack.

That type of information doesn't change often, so answers are useful over a long period of time. It also means you can collect it throughout the year, over the course of multiple campaigns.

Answers That Are Useful Only in the Short-Term—for a Few Weeks or Months

These include, for example, whether prospects are attending marketing forum this fall and whether they're interested in attending an email marketing meetup. This kind of information is incredibly valuable.

However, such questions need to be asked close enough to the event to ensure respondents are certain they're going, but not so close that you don't have enough time to act on their responses.

5. When Analyzing the Responses

Get it all right up to this point... and you can still stumble when it comes time to interpret the results. Consider the following four issues.

A. Whether to Report N/A, Not sure, and Don't know Responses

Generally, such answers aren't meaningful, so it's best to remove this noise from your reported results.

However, it can be telling if, say, most or a plurality of respondents select those types of answers. That can signal that the technology, tactic, product, or whatever else you're asking about has low awareness, which can be interesting in and of itself.

It could also signal that your question is confusing.

B. Look for Opportunities to Simplify the Story

Just as you don't have to report your N/A answers, it's OK in some cases to roll together responses to tell a cleaner story.

For example, say you asked respondents to respond to a statement using a 5-point Likert scale of (1) Strongly Disagree, (2) Disagree, (3) Neither Agree nor Disagree, (4) Agree, and (5) Strongly Agree. In some circumstances, it might make sense to combine the two disagree answers and the two agree answers when reporting results.

C. Reporting Statistically Significant Results Across Segments

Particularly with demographic questions, brands often provide a long list of answer choices because they want more granular data. For example, they might ask about company size and provide lots of choices, such as Fewer than 10 employees, 11-25, 26-50, 51-100, 101-200, 201-500, 501-1,000, 1,001-2,000, ,2001-10,000, More than 10,000.

Reporting those responses is fine; it gives your audience valuable perspective on your respondents. However, sometimes, brands then try to report how each of those groups answered other questions. Depending on your brand's audience and how the survey or poll was fielded, you might have a relatively small number of respondents in some of those buckets—too small for the results to be meaningful.

In many instances, it makes sense to combine some answer choices. For example, in past surveys, I've rolled together respondents by company size into two buckets: 500 or fewer employees and More than 500 employees. That dividing line created two groups that were fairly evenly powered and provided interesting insights about the differences between what smaller companies and larger companies were doing.

D. Understanding Intent

Just as people are bad at remembering things that happened more than a year ago, they tend to be bad at predicting what they and their organization will do in the year ahead.

In my experience—and depending on what you're asking about and how much effort, cost, and buy-in is required—fewer than half of respondents follow through on what they say they'll do in polls and surveys.

Needless to say, that's totally fine. Your respondents are answering to the best of their ability. That said, when you report the results, you shouldn't overstate the results of such intent-related questions.

6. When Repeating a Question, Survey, or Poll

Collecting data on the same question over and over (e.g., every year) is powerful, but there are a couple of issues to be mindful of.

Be mindful of changing previously asked questions and their answer choices

Doig so will likely render historical answers useless for comparison or trend purposes. Even changing an introductory statement or clause can change the responses so much that you can't compare them to past responses.

That said, if you've identified a serious flaw in the wording of a previous question, don't hesitate to reword it so you get more reliable answers next time.

Similarly...

Be mindful of controlling your audience

For example, if you promoted a poll question to your email and social audiences last year, but then this year you work with partners to have them also share your poll question, the poll results could be materially different because of that audience change.

That's not to say that expanding your audience is bad, but don't disregard that change when analyzing results.

As Privacy Protections Strengthen...

Companies need more ways to collect information about their customers and prospects so they can understand their audience better and create more relevant experiences.

Asking your audience questions through forms, surveys, polls, and other progressive profiling mechanisms is a highly valuable way of staying close to them and serving them better.

-----------------------------------------------------------------------------

If you need help with your email, web site, video, or other presentation to promote your company, product, or service, please give me a call at 440-519-1500 or email me at john@x2media.us

Until next month. . . .remember. "you don't get a 2nd chance to make a 1st impression." Always make it a good one!!