AMA: Automation Anywhere Senior Vice President, Product & Solutions Marketing, Claudia Michon on AI and Product Marketing

Generative AI is a tool, an assistive technology. We all need to keep this in mind, especially leaders who erroneously think how they can replace their people with this tech. Why is this the case? The tech (today) cannot do one critical thing - creative, strategic thinking. It's dumb. It can serve data and knowledge, but it doesn't have wisdom. What makes the best PMMs is wisdom, having the ability to harness past experiences, customer needs, and sales requests, understand the market and the technical components of a product - and use all of that to make decisions, craft stories and bring something new to life. Keeping that in mind, I think of generative AI as helping with the microtasks that slow our work. Get a great tool that's flexible to grow with your team. There are quite a few out there. We have a tool that we use company-wide that's sanctioned by our IT team. In addition to Chat GPT there is also Glean, Jasper, Writer - and many more. You can create your own "GPTs" - essentially building knowledge bases with relevant inputs and data and training them to meet your needs, like giving them skills. I have a list of GPTs my team has built or plans to build. You're going to have to experiment here, so be patient. Example GenAI for PMM Use Cases: * Research a competitor's product lines or messaging * Turn video transcripts into blog posts * Create engaging demo scripts from step-by-step how-to's * Email nurture series generation based on core messaging and target persona * Create customer case studies from interviews * Create a GPT for a specific persona and rapidly generate value prompters and objection-handling documents * Brainstorm product names, taglines, or website headlines Will the tech get smarter? Of course. I recently heard a great quote from Seth Godin - "it's clear that AI is as dumb today as it will ever be again tomorrow." We all must keep up and become expert users to elevate our skill sets and remain relevant contributors to the success of an organization. Have fun!

Yes! It's really helped me speed up my work. I have a template that's attached to this AMA that gives examples of how to do exactly this. The rule of "garbage in, garbage out" totally applies to generative AI tools, models, and LLMs. With RAG, you can enrich your GPTs with relevant data for your customers, products, industry, and competitors to tune a model to become more relevant. Example source content: * Persona cards * Messaging frameworks * Sample content (blog posts, customer interviews) * Industry reports * Analyst reports * Brand tone guidelines * Press releases You can add as much data as you need to make your GPTs as informed as possible. Just make sure it's the right content.

Massively. Automation Anywhere is a generative AI-powered platform, so my team and I must constantly read what's happening in the market to keep up. It has to be a range of sources, from data science influencers to the latest launches and campaigns. On an operational level, we are building GPTs (or Apps) that support the basic PMM tasks we'd love to offload. I'm about to update our OKRs to include the development of GPTs to support various functions. My leads work with our IT teams to add these to our company's sanctioned platform so we can distribute them for broad use. We have a few that are currently in production and are rolling more out every week. Using these tools gives us more headspace to think about how to do things more creatively, vs. just getting things done. We're still a bit early so the big benefits haven't been realized yet, but I'm super excited for how the PMM role is going to become much more strategic.

Every PMM needs to become AI-literate. If you're an IC, you need to think about how to use AI to help you scale. If you're a manager, you need to think about how it can help your team scale. If you're a senior leader, you need to think about how it can help your company scale - and even transform your industry. Today most of the top martech platforms are building generative AI tools into their functionality, but that's not enough. PMMs should always be reading about how the tech works, how it's evolving, and think about ways to apply it to their work. IT review and procurement processes don't move fast enough to keep up so you need to stay ahead of the game. Prompt engineering is a critical skill that is going to help PMMs accelerate their work. I think PMMs are naturally adept at this as it requires consideration of the audience (GPT) to get the desired outcome. Starter prompt templates for various PMM tasks are all over the internet. But they aren't great and you'll need to tune these to get them to spit out relevant, quality content. It's also to ensure you're up to date on the limitations of the tech. It's not a replacement for PMM work. It's assistive. Be wary of hallucinations and always double-check the outputs. And also be careful that you aren't putting your company's IP into the public sphere. That could get you into deep water with your legal team! Helpful Tip and Resources * Follow Andrew Ng on LinkedIn for thought leadership on the tech * Subscribe to the TLDR newsletters for Marketing and AI * Podcast: The Artificial Intelligence Show * Watch events and read the blogs of the top players in the space (OpenAI, Google, AWS, Microsoft, Nvidia, IBM, Anthropic) * Use AI to master AI - chat with ChatGPT and ask questions like, "Explain RAG to me like I'm in high school" or "Give me 10 ways to use generative AI for [insert use case] LinkedIn Learning Courses LinkedIn Learning has a couple of courses on offer. these are great to get a baseline, but remember, the tech is changing so don't expect them to be up-to-date: * Securing the Use of Generative AI in Your Organization * Introduction to Prompt Engineering * Introduction to Large Language Models * Getting Hands-on with GPT 4

As I answered in another question, generative AI is an assistive technology. It's basically in the toddler stage. It's terrible at critical thinking and creative problem-solving. That is the key skill set of a good PMM. I think the tech is going to be an incredible accelerant for PMMs. But I will say that all knowledge work is going to change dramatically as many of the basic tasks will be picked up by AI. I don't know what this will mean for the future of knowledge work. There are many predictions from the optimistic to the doomsday. For the near future, I'm in the optimistic camp. But as a senior leader, I excitedly welcome more help in the basic, time-consuming work that isn't really high value for me. This means that more entry-level roles are going to change faster. Anyone who is currently in college studying marketing (or any knowledge work profession) needs to ensure AI is a core part of their curriculum or they risk not finding a role when they graduate. In the not-so-near future, I'm not sure if PMM will exist. I'm not sure if lawyers will exist. I'm not sure that coding will be a thing or video post-production. It can be scary, but it's also an opportunity and imperative to grow and expand beyond what we think is possible today.

Dropping in some resources from a previous response! Helpful Tip and Resources * Follow Andrew Ng on LinkedIn for thought leadership on the tech * Subscribe to the TLDR newsletters for Marketing and AI * Podcast: The Artificial Intelligence Show * Watch events and read the blogs of the top players in the space (OpenAI, Google, AWS, Microsoft, Nvidia, IBM, Anthropic) * Use AI to master AI - chat with ChatGPT and ask questions like, "Explain RAG to me like I'm in high school" or "Give me 10 ways to use generative AI for [insert use case] LinkedIn Learning Courses LinkedIn Learning has a couple of courses on offer. these are great to get a baseline, but remember, the tech is changing so don't expect them to be up-to-date: * Securing the Use of Generative AI in Your Organization * Introduction to Prompt Engineering * Introduction to Large Language Models * Getting Hands-on with GPT 4

Critical thinking and story-telling are the two top skills I look for in a PMM. It's way too hard to put that into an algorithm (today). In addition to that, I think every employer will be looking for AI to be a part of a candidate's tool kit. Literacy in the latest tools, comfort with prompt engineering, and a bit of an engineer's mindset are going to be characteristics of the PMMs of tomorrow.

Totally a team productivity play. That's where most companies are right now. How to do work faster. Soon it will become how to expand our ability to innovate and after that, we'll be spinning out new market opportunities! This year, each team member will have an OKR related to creating an AI tool to help them with their work. We have support from our IT organization to do this, which is a great help!

Oh boy, this is a fun one. First, it's important to note that my team has a really strong working relationship with Product. It's not always the case in every company, and I think it's fair to those who don't that I provide this context. I do my own reading and research. I need to know enough to understand what good looks like and what's not meeting the bar for differentiated innovation. You can't do that if you're just sitting around and waiting for Product to hand you a list of features. I am also incessant with asking PMs questions about everything. I am constantly asking clarifying questions and challenging assumptions - both to learn and to stretch the sometimes-constrained thinking of my peers. Before you reach out, come with a baseline level of understanding to help the conversation be as productive as possible. Do a lot of competitive research. It will be non-stop because the market is changing so rapidly. Become a source of market knowledge to your PM counterparts. Draft potential stories you can tell in market. Come up with new ways of packaging or branding your AI offerings that might seem unconventional. Push the limits and be radical. It's more fun that way and you'll be able to work with Product to find a happy medium between what's explicitly in the code and the aspirational marketing fluff.

I want to challenge the statement "Customers don't understand what AI can do." Sure, some don't, but many in the automation space do and in fact, we are learning quite a lot from the initial use cases we see happening across our accounts. Our CS team is always feeding us new ideas so we can integrate them into our product roadmap and capture them in our messaging. There is certainly a lot of opportunity being left on the table as the bounds of AI are relatively limitless (and the limits keep getting pushed out!) In those cases, instead of asking for the answer, we ask for the approach to getting to the answer. For example, instead of asking "What are you using generative AI for?" we ask "What are the problems you are trying to solve?" Here is a set of sample questions to include in your customer research. General Research Questions * What are the problems you are trying to solve today? * What are the biggest barriers to solving those problems? * What approaches have you tried - why have they failed/succeeded? * How do you view our platform/product as being able to solve your problems or meet your needs? * Where do we fall short? * Where do we exceed expectations? * What functionality or solutions would you love to see built into our product in the near future? Once you have a baseline without any assumptions, you can follow up with more specifics on generative AI usage. Generative AI Usage Questions * How are you using generative AI to solve your problems today? (can you get specific on use cases?) * How is it helping/what are the outcomes? * What are the limitations you are seeing? * What innovation or training would help overcome the limitations? * How is it being adopted? What are the barriers to adoption? * What is the sentiment around the use of this tech? * How do you see it helping you do your job better? Your team? Your company? * How much are you spending on the tech? How much do you wish you could spend? Who manages this? The list could go on, but my point is to never dive right into the tech. Always start with the problem/opportunity and better understand how to solve it for your customer.

Our superstar enablement team is in the process of piloting a custom tool that will just get better over time! We're also leveraging new features of our LMS and enablement platforms. Whatever tool you use, remember to crawl, walk, and then run. You'll learn a lot about seller behavior and adoption to iterate on your next version. 1. Search: Enable sellers to more intuitively find a piece of content. 2. Discover: Create a knowledge base for sellers to search for on-demand answers. 3. Evaluate: Automate sales certifications - Have genAI review pitch transcripts to ensure sellers know the talk track. 4. Create: Enable sellers to auto-generate customer-facing content (right now this is too high-risk!)

When you have feature parity, it sometimes comes down to execution. Here are some levers you can use to break away from the pack: * "Ours is just better." Explain why it's easier, more accurate, more flexible, more cost-effective, faster, etc. Find ways to shine a light on why your flavor is the right choice. * Customer success. Nothing is more powerful than proof of adoption and examples of use cases and success. * Give it a great brand and positioning. Make up a new category. The world of AI is the wild west and basically anything goes. It's a PMMs dream right now. * Position partnerships and ecosystem. The extensibility and interoperability of features is what makes it even more valuable. Tout a library of integrations, joint investments, and solutions with strategic technical partners. * Make noise. Get your message into market and establish a stronghold in hearts, minds, and desktops.