Can Generative AI Replace Marketing & Enablement? Get Ready To Drive UBER...

I’m WAY LATE to the Generative AI and ChatGPT “game”.? I mean big time late, like getting to dinner when dessert is being served late.? Probably something to do with being a Gen X'er.? Heck, I was late to texting by a decade.? I remember this young kid I knew back in the day in NYC who was constantly texting his dates on a flip phone while walking down Park Ave.? I remember asking myself, “What in the world is this guy doing with his phone?”? Anyway, here I am, finally getting on the bandwagon.?

You see, my old friend Dave Wilt and I reconnected recently after many years, and he just published this really thought-provoking article exploring whether or not OpenAI could possibly replace a B2B marketer.? My other dear friend Sundar Nathan has written many articles and books on AI and has posed similar questions.

Was this really possible?? I wanted to find out for myself.? So I finally signed up for ChatGPT, and using just the FREE version, my life - and my life’s prospects for that matter - completely changed in about 30 minutes.? I just need to start driving for UBER right now.? Don’t pass go, don’t collect $200.? Just…UBER.? Or maybe Starbucks.? You can’t “AI” a latte into a frothy existence, can you?

I really put this "AI bot thing" through the wringer.? I went pretty deep on it (at least by my neophyte standards).? I wanted to see if I could get it to create a complete messaging and content stack for a B2B software company; pretty much all the content that would go into a sales playbook, all the tools for sellers and SDRs/BDRs, as well as all customer-facing content and even advertising.? Essentially the combined job of a high tech Product Marketer AND a Sales Enablement manager for an entire month.? It’s a pretty lengthy read, just warning you. The AI brain has lots to say, apparently.? And I wanted to be thorough, at least enough to have a definitive answer to the question posed by Sundar and Dave.

I put myself in the shoes of a newly hired product marketer working for a vendor selling a real-time analytics database.? I wanted to be pretty specific about the technology niche (not being too generic) in order to see how well the engine could handle the highly nuanced messaging of real-time analytics databases.? What I got back was truly eye opening.

The first thing I did was get "it" to help me target the right ICP (Ideal customer profile) for a real-time analytics database.??

I asked,?

“What are the specific types of companies that will most likely need the unique capabilities of a real-time analytics database in 2023?”??

What I got back was a shockingly accurate response:

"Real-time analytics databases are most useful for companies that need to process and analyze large volumes of data in real-time or near real-time, and make decisions based on the insights gleaned from this data. Some specific types of companies that may require the unique capabilities of a real-time analytics database include:

1. E-commerce companies: E-commerce companies rely heavily on real-time data to track customer behavior, optimize pricing, manage inventory, and personalize the customer experience.
2. Financial services companies: Financial services companies, such as banks and investment firms, require real-time analytics to monitor transactions, detect fraud, manage risk, and make investment decisions.
3. Telecommunications companies: Telecommunications companies use real-time analytics to monitor network performance, detect and troubleshoot issues, and improve the customer experience.
4. Healthcare organizations: Healthcare organizations require real-time analytics to monitor patient health data, manage hospital resources, and make clinical decisions.
5. Online gaming companies: Online gaming companies use real-time analytics to track player behavior, optimize gameplay, and prevent fraud.
6. Advertising and marketing firms: Advertising and marketing firms rely on real-time analytics to track campaign performance, optimize targeting, and personalize content.

Overall, any company that requires real-time data processing and analysis to make informed decisions can benefit from the unique capabilities of a real-time analytics database."

Not only is this list comprehensive and 100% accurate, notice how the AI engine decided on its own to add "color commentary" about exactly why each customer segment would benefit from real-time analytics, and how each would use it. Score one for the AI bot thingy.

Next, I wanted to create a list of target personas so my sellers would know exactly who to prospect into.? But now I explicitly wanted a PRIORITIZED list…

…so I said to the machine,

“Compile a list of buyer titles for buyers of real time analytics databases, in order of how important they are to the decision to acquire a real-time analytics database.”

I visibly deflated as I read the response:

“Here's a list of buyer titles for buyers of real-time analytics databases, ordered by their level of importance in the decision-making process:

1. Chief Technology Officer (CTO)
2. Chief Data Officer (CDO)
3. Director/Manager of Analytics
4. Director/Manager of Data Science
5. Business Intelligence Analyst
6. Data Engineer
7. IT Director/Manager
8. Chief Financial Officer (CFO)
9. Chief Executive Officer (CEO)
10. Chief Operating Officer (COO)”

First, notice how the AI rewrote my query to read better.? Thanks.? Now, about the list itself - ?pretty accurate.? I would have ranked “Data Engineer” a bit higher, but otherwise it’s spot on.? Now I wanted to know what the AI thought about who would typically be the Champion in these opportunities, and who would be the Economic Buyer.? This is very important when guiding sellers on who to target within enterprises and even mid-sized companies.

So my next question was,

“Which buyer title is typically the "champion" in a real-time analytics opportunity, and which buyer title is typically the "economic buyer"?

Well you guessed it… “It’s a bingo!”

“In a real-time analytics opportunity, the 'champion' is typically the Director/Manager of Analytics, as they are the primary users and stakeholders of the solution. They are responsible for evaluating and recommending the best solution for their team's needs.

The 'economic buyer,' on the other hand, is typically the Chief Financial Officer (CFO) or Chief Operating Officer (COO). They are responsible for the financial aspect of the purchase decision and have the final say on whether the investment in the real-time analytics database aligns with the company's financial goals and objectives.”

Okay, so now I know who my Champion will most likely be in these deals.? Next, I wanted to figure out which technical requirements are most important to this specific Champion persona, so I can arm my sellers (not to mention my product team) with this valuable information.?

Such interesting work this is, feeding this engine thought-provoking and well-worded questions!? Is this the next great skillset people will put on their resumes? Understanding how to talk to this thing in just the right way to get it to share its deepest, darkest secrets?

Back to the task at hand... I asked - or rather instructed…

“Compile a list of technical requirements important to the Director/Manager of Data Analytics when evaluating real-time analytics databases.”

Now when you read this response, keep in mind I did NOT ask the engine to order these items in terms of their importance in the decision-making process.? But that’s the cool (not cool, SCARY, why do I keep saying “cool”?) thing about the AI brain thingy, it pretty much does that kind of stuff anyway without needing to be asked.

"Here are some technical requirements that Director/Manager of Data Analytics may have when evaluating real-time analytics databases:

1. High performance: The database should be able to process large volumes of data in real-time or near-real-time with low latency and high throughput.
2. Scalability: The database should be able to scale horizontally or vertically as data volumes grow and the number of users increase.
3. Data modeling: The database should support flexible data modeling and schemaless data structures to handle a variety of data types and formats.
4. Query capabilities: The database should have powerful and efficient query capabilities to allow users to quickly retrieve data and perform complex analysis.
5. Integration: The database should be able to integrate with other systems and applications within the organization's technology stack.
6. Data security: The database should have robust security measures in place to protect sensitive data and ensure compliance with data privacy regulations.
7. Backup and recovery: The database should have reliable backup and recovery mechanisms to prevent data loss in the event of a failure or outage.
8. Management and monitoring: The database should have tools for managing and monitoring database performance, including monitoring for slow queries, identifying bottlenecks, and managing resources.
9. Availability: The database should have high availability and be able to operate continuously without interruption.
10. Support and documentation: The database vendor should provide excellent support and documentation to assist with installation, configuration, and troubleshooting."

I would say the completeness of this list is 100% and the ranked order is around 80%.? Yes, Carl, I would have ranked “Availability” a bit higher in the list.? Otherwise it’s spot on.

As I pondered what I had just generated (or rather, what I had asked The All Knowingness to generate), it occurred to me that back in "normal" times it would have taken a room full of tenured Marketing Directors eating a lot of catered food at a pretty nice hotel banquet center at least three 18-hour days to come up with (and then actually agree on) the sheer volume and detail that popped up on my screen in like 3.6 seconds.

The artificial mind has been pretty spot-on so far in terms of high level requirements, buyer personas, and that kind of thing.? I needed to keep going, keep getting more nuanced. Could it discern the discrete USE CASES of a real-time analytics database, but within a specific sub-segment of the Financial Services industry, for example?? I chose something fairly obscure but nonetheless quite relevant to my sellers, Portfolio Management.? After all, Portfolio Managers typically have lots of yachts and jets and Rembrandts lying around, not to mention stacks of cash to spend on cool technology, so let’s go with them.?

I tried to get fancy by asking a "compound" question:

“What are the specific use cases for real-time analytics databases in the portfolio management segment of the financial services industry? And for each of these use cases, please list the best qualification questions that a seller should ask a portfolio manager to ascertain their level of interest in a real-time analytics database solution.”

Despite the proficiency of the engine in the earlier examples, I was definitely not expecting anything as accurate and detailed as what I got back:

"Real-time analytics databases have a variety of use cases in the portfolio management segment of the financial services industry. Here are some specific examples:

Real-time portfolio management: Portfolio managers need to be able to monitor their portfolios in real-time to make informed investment decisions. A real-time analytics database can provide instant access to portfolio data, including performance metrics, risk analysis, and asset allocation.

Qualification questions to ask a portfolio manager related to Real-time portfolio management:

• How often do you need to access portfolio data?
• What are the most important metrics you track when managing your portfolio?
• Have you encountered any challenges with accessing portfolio data in real-time?

Risk management: Managing risk is a critical part of portfolio management. Real-time analytics databases can help portfolio managers monitor risk factors, such as volatility, liquidity, and credit risk, in real-time.

Qualification questions to ask a portfolio manager related to Risk management:

• How do you currently monitor and manage risk in your portfolio?
• What specific risk factors are most important for your portfolio?
• Have you encountered any challenges with monitoring risk factors in real-time?

Performance analysis: Portfolio managers need to be able to analyze portfolio performance to identify areas for improvement and make informed investment decisions. A real-time analytics database can provide instant access to performance data, including benchmarks, returns, and attribution analysis.

Qualification questions to ask a portfolio manager related to Performance analysis:

• How do you currently analyze portfolio performance?
• What metrics do you use to evaluate performance?
• Have you encountered any challenges with accessing performance data in real-time?

Alerting and notification: Portfolio managers need to be able to receive real-time alerts and notifications about significant events or changes in their portfolios. A real-time analytics database can provide instant alerts for events such as changes in asset allocation, breaches of risk thresholds, or significant market movements.

Qualification questions to ask a portfolio manager related to Alerting and notification:

• How do you currently receive alerts and notifications about your portfolio?
• What specific events or changes would you like to be alerted about in real-time?
• Have you encountered any challenges with receiving real-time alerts and notifications?

When speaking with a portfolio manager about a real-time analytics database solution, it's important to understand their specific needs and pain points in managing their portfolio. By asking thoughtful qualification questions, you can better understand how your solution can meet their needs and provide value."

These use cases aren't just 100% accurate, they're RANKED in order of how I would tell my sellers to go after this specific market segment.? The qualification questions are thoughtful and well-worded. This content would have taken me at least an hour to create. And that's assuming I could get our top Financial Services seller on the phone to vet and flesh out the material.

As I pondered the time savings, it occurred to me that B2B Marketing and Sales teams are increasingly measured on productivity and efficiency.? Will people actually lose their jobs? There's not enough data on this yet, but we've all heard that saying, "do more with less", right? If this kind of generative AI can make a typical B2B marketer or enabler 2-3x more productive, well….do the math. I built a good size chunk of a Sales Playbook in 20 minutes.? Yeah, I need a couple more days to fine tune it and make it vendor-specific, for sure.

Wait....vendor specific?? What’s stopping me from using actual vendor names in my questions?? it’s not like there’s some unspoken “AI law” against asking for what used to be secret squirrel information, right?? So I asked a question that I really, really hoped it wouldn't - or couldn’t - answer correctly (it appears to always answer with something).? A question that one might think is the last bastion of hope and self-respect for the high tech marketer or enabler.? A question a salesperson asks when they're competing in a deal and they want to know what to do.? A question that goes something like this:

“I'm a salesperson selling Apache Druid, and I'm competing against Amazon Redshift. What are the points I should highlight with my prospect that give me a clear advantage to win the deal?”

And it was finally at that precise moment, as I read the AI-generated response, that I witnessed the firepower of this fully armed and operational AI technology, and what it would mean for entire swaths of the knowledge worker universe.

"As a salesperson selling Apache Druid, there are several points you can highlight to differentiate your product from Amazon Redshift and give yourself an advantage in winning the deal. Here are some key points to consider:

1. Speed: Apache Druid is designed to deliver fast query response times, even when working with large datasets. This can give you an advantage when competing with Amazon Redshift, which may struggle to keep up with complex queries or large volumes of data.
2. Scalability: Apache Druid is highly scalable, which means it can handle large amounts of data and can grow as your needs change over time. This can be an advantage over Amazon Redshift, which may require more time and effort to scale up or down.
3. Real-time Data: Apache Druid is optimized for real-time data ingestion and query processing, which makes it a good choice for applications that require up-to-date insights. This can be an advantage over Amazon Redshift, which may have delays in data processing due to its batch processing approach.
4. Flexibility: Apache Druid is designed to be flexible and can work with a variety of data sources, including structured, semi-structured, and unstructured data. This can be an advantage over Amazon Redshift, which may be limited in its ability to handle certain types of data.
5. Cost: Apache Druid is an open-source project, which means it can be a cost-effective option for organizations that want to avoid paying for proprietary software. This can be an advantage over Amazon Redshift, which may require significant licensing and infrastructure costs.”

By highlighting these points, you can make a compelling case for why Apache Druid may be a better choice than Amazon Redshift for your potential customer's needs."

And no, Dave, these aren't what I would call “weakly-vetted” AI responses.? These are the exact points I would have highlighted, in order of importance, and perfectly worded. What this means, my friends, is that all that work I said that needs to be done to “polish” the above playbook content to make it vendor specific?? Poof.? The vendor-specific Sales Playbook is now 100% done in about 8 minutes.

But then it occurred to me; who even needs a playbook now?? And who even needs a Competitive Intelligence team (or Wikis stuffed with competitive info)? Now I can just tell my sellers to ask the AI thing what to do in any situation, how X competes with Y, what the weaknesses of Z are, and they’ll get their answers far faster and probably even more accurately than looking it up in some outdated playbook or battlecard or online content management tool that I slaved over for two millennia.

As I wondered how far down the rabbit hole I was willing to go, I couldn't help think about how BUYERS themselves would start using AI to cut sellers out of the process altogether. After all, if I was a buyer looking to understand product differentiation, wouldn't I just ask the AI,

"If I'm a portfolio manager who needs to do real-time portfolio management, alerting, and risk management on huge volumes of event-driven data, which product should I use, Apache Druid or PostgreSQL?"

When I read the fluid and artful response that landed on my screen like a laser-guided precision munition, I realized that the primary role of B2B sellers might sadly become sending prospects dueling, carefully worded questions that they would then feed into their AI consoles in order to (hopefully) generate the desired response. All Quiet on The Western AI Front. Endless exchanges of carefully targeted digital artillery barrages across No Man's Land. Lovely.

So let’s review so far.? The job of product marketers and enablers will shift completely from content creation to content compilation, at best. Or maybe the job becomes COMPETING with AI and proving how inaccurate it is (I've shown it really isn't). Can you just see it? A Product Marketer publishes a sales tool and a rep sends back the AI-generated version, politely asking, "which one is right?"

Let's quit picking on Product Marketing & Enablement. Let's spread the wealth around. Will other teams be equally impacted? With that question in mind, I turned my twin-barreled AI sights unflinchingly towards the Digital Marketing Team. It's nothing personal, folks.? As my Grand Dad used to say, “Why should you have all the fun while I’m out to pasture?”

So I unsparingly asked,?

“What are some good ideas for digital advertising tag lines for real-time analytic databases?”

Now remember, digital marketers spend an INORDINATE amount of their available brain power coming up with these tag lines.? I’ve been on my share of Zoom brainstorming sessions where it gets really loud and contentious because nobody can agree on much.? There’s just a lot of, “Yeah, let’s meet again on this next week”, and it just goes back to committee.? Now we can all just say, “Well, the AI engine said this, so we should probably just go with it.”

“Here are some ideas for digital advertising taglines for real-time analytic databases:

1. "Stay Ahead of the Curve with Real-Time Analytics"
2. "Real-Time Insights for Real-Time Results"
3. "Make Smarter Decisions, Faster"
4. "Instant Insights for Instant Impact"
5. "Transform Your Data into Actionable Intelligence"
6. "Unlock the Power of Real-Time Data"
7. "Accelerate Your Business with Real-Time Analytics"
8. "Real-Time Analytics for Real-Time Business"
9. "Data-Driven Decisions, Delivered Instantly"
10. "Experience the Power of Real-Time Insights"

Now I’m no digital marketer, but I will tell you I really like a couple of these.? Do they already exist?? They might, not sure.? But in my non-digital-marketer's opinion, “Instant Insights for Instant Impact” is isn't bad at all.? “Real-Time Analytics for Real-Time Business” is pretty good also.? Let’s just say nobody would get fired for going with five or six of these. ??

It seems no team is beyond the AI Grim Reaper's reach. Where would the line be drawn? That's just it. This stuff OBLITERATES lines. It takes no prisoners. It leaves no knowledge realm unscathed. It is mercilessly ruthless in its destructive dehumanization of any and all creative endeavors. In fact, it risks relegating future creative innovation to a prison of past collective consciousness. If you think movies are formulaic today, just wait until AI starts writing them based on past examples. It makes one just want to throw up one's hands and say,

After reflecting deeply on what I had just learned about the tectonic shifts this technology will bring about, I somehow (albeit briefly) oscillated back to a surprisingly optimistic point of view. It occurred to me, thankfully, that the “human element” is still the purview of - yes - us humans!? I recently wrote an article about how Customer Intimacy is the most important Discipline of Excellence for B2B companies.? I talked about how we need to show more empathy and admit when we’re wrong when our solutions fail in order to build lasting and meaningful customer relationships.? THANK GOODNESS this kind of empathy building is simply not something that a cold-hearted, logic-oriented AI engine could possibly come up with!

And then I did it.? Yeah, I just did.? I just had to know.??

I asked:

“What is the right thing to say when a customer calls up and says that your software has completely failed and it has caused them huge problems?”

And the thing actually responded with:

“When a customer calls up and informs you that your software has completely failed and caused them significant problems, it's essential to acknowledge their frustration and offer empathy. Here's an example of what you can say:

"I am sorry to hear that you are encountering significant problems. I can understand how frustrating and inconvenient this must be for you. We take all complaints seriously, and I want to assure you that we will do everything we can to resolve this issue promptly. Can you please provide me with more details about the problem you are experiencing so that we can begin investigating and find a solution as soon as possible?"

By acknowledging their frustration and showing that you take their complaint seriously, you can help to deescalate the situation and reassure the customer that you are committed to finding a solution. It's also important to actively listen to the customer's concerns and take ownership of the problem to show that you are accountable for the software's performance.”

If AI is better at empathy building than most sellers and customer service reps, where does that leave us bags of mostly water for pretty much ANYTHING!!!??? UBER may be in all of our futures.? That is, until Waymo takes that away too.

Seriously though, as a B2B Enablement Professional, where do these stark realizations leave me? Where do they leave anyone trying to help sellers be more effective? Content creation and even compilation are over. Let's just face it. Whitepapers are over, as are Solution Briefs (what buyer would even read them now anyway). If the FREE version of early AI is already this good, what will the PAID version look like in 3 years? Hard skills won't be as relevant either, as various, discrete tasks that required specialized training before will now be guided by AI or handled by it entirely. Soft skills will be the last bastion, the last line of defense. Enablement is actually needed now more than ever to help sellers retain and sharpen the human element in how they connect with buyers in the ever-shrinking window when a buyer might even contemplate connecting with them. Enablement can also help make selling fun and, through gamification, events, challenges, and call blitzes, very rewarding on both a personal and financial level.

Humans are humans for a reason. We're fallible, imperfect things. We're not machines. We don't want to become machines, or live in a world where interacting with machines is the rule rather than the exception. We crave human interaction. Human conversation. Human imperfection. Human drama. My life's passion has been about how to train sellers to have better, more intimate, more empathetic, and more meaningful conversations with buyers. Buyers still want to have those too, because, yes, they're humans as well (surprise!), and they want to be connected with in an authentic, transparent, and interesting way.

We all crave a level of meaning and connection that AI will never replace, not even if the video bots get really, really good at human inflection and interpreting gesture and eye contact and tone. Tyrell was right to build a failsafe into the Nexus-6 series. Are we smart enough to do the same? Let's at least build a failsafe into how we train our sellers. Let's show them the value of creative, inter-personal connection (especially the thoroughly-phone-addicted Gen Z'ers). If we don't, we risk future generations of sellers being completely disintermediated from the buying process, to the point where they're just taking orders (if even that). Instead, let's live in a world where sellers and buyers still dance the dance, and steak dinners are still maybe a thing, albeit on rare occasion.

And let’s face it, AI could never put on a real life, physical trade show.? Trade shows are fun.? Not really money well spent, but super, super fun.? It's where we'll all go for refuge as AI creeps deeper and deeper into our lives. And late one night after the show ends, we’ll all be sitting around in a bar somewhere, and you know what we'll be saying...“Yeah, I used to be in B2B sales.... it was a tough racket…”.