·
13 minute read
As the dust starts to settle from the meteor-sized impact AI has had on marketing, teams are figuring out how to make this powerful new tool work.
In Part 1 of this series, we revealed how some of our late AI adopters reconciled with their reluctance and grew to embrace AI. In Part 2, we covered how we balance AI with human smarts and discussed real life use cases for AI in day-to-day workflows.
But there's still a critical layer to address: How could we turn all of this into smart, scalable company policy? If you’ve been experimenting with AI and want to standardize and optimize how you use it, read on … or watch the video below!
Meet Our AI Visionaries
Beth Carter is Clariant Creative’s founder and Chief Strategist and has led the organization through many changes, both big and small. Caitlin Castevens is our Director of Client Services and is focused on ensuring Clariant Creative delivers the best possible strategy and service to our clients.
We got their conversation on video, moderated by our Inbound Marketing Associate, Liza Park. Check it out, or read the transcript below, which we’ve edited for clarity.
“When it comes to using AI for any type of agency deliverable, what concerns are top of mind for you?”
Beth
So, a few things come top of mind when we're looking at both the AI tools to use and the AI processes to use.
As a business owner, the first is, “Are we being consistent in the way everybody across the team is using these tools?” I might have my way of doing things and Caitlin might have her way of doing things and someone else on the team might have their way of doing things within any of the AI platforms and tools out there. And as the agency owner, I want to make sure that we get out of those little silos and have some standardized processes across the agency so that everyone knows, “For this type of project, this is how we're using this particular tool.”
To my mind, that's what helps make the solution scalable and sustainable and ensures that the results we're generating are consistent.
Caitlin
As a people manager, I'm thinking about how this starts at the human level. People are so worried about the technology. They're worried about things replacing their jobs. They're thinking about how they've always done things and having to change the way they're doing their work moving forward.
I think sometimes we forget about that — about how people feel and what they're scared about and allowing team members to unpack some of those fears. If you can name it, you can tame it. I always say that, right? I think the same is with AI. If we can help people find some confidence in navigating something new, then that will help reduce some of the worries and fears that are the stories that people are telling themselves.
Beth
The flip side of that is it's also important to make sure we're adequately bringing these topics up with the clients that we serve as well. Our clients don't know what they don't know, and they have no idea necessarily how the sausage is made. And nine times out of ten, that's fine. That’s perhaps more detailed than a client would ever care about.
That said, AI has the potential to change a lot of that. And one of the things that we're working through right now internally as an agency is how do we broach these conversations with our clients? What level of information do our clients want? What level of AI usage do our clients want us to be doing?
There's a whole lot of conversation that we're just starting to dip our toes into right now. I think those are going to be very important conversations. We're actively working on how to approach and maintain and hold these conversations right now. I don't know that we've got any clear-cut answers on what that looks like. I suspect we'll have different approaches with different clients, because all our clients are at different points in their journey. And so we want to be really transparent about how we're leveraging AI to help their own journey and growth toward revenue.
Caitlin
And there are a lot of debates in the AI space about should you be AI first or AI forward? And Paul Roetzer from SmarterX.ai and PR 2020, he's really advocated for this AI-forward approach because AI shouldn't lead the process, it should support it. A person still needs to start the conversation, they still need to make sense of the output and ultimately shape the final result.
Something that we've been also talking about is how to share different approaches of what our team is doing and, our clients are asking, how are you guys experimenting? How are you testing? How are you using it in a way that is not too risky or not breaching security or how are you using it to brainstorm or refine a thought or a strategy or disseminate and analyze information that's public-facing. We do talk a lot about priming just as much as we talk about prompting, which means giving AI the right context, background and nuance, just like you would a new hire. So this concept of being AI-forward, instead of AI-first, brings us back to our values of being genuine, right? And being meticulous.
Beth
And that also speaks a lot to the strategy that we bring, right? Where we bring so much value — and hopefully AI helps multiply that value — is the care and effort we put into priming the model before we're even ready to start talking about an output; building out the context for the tool, and the framework for how we want the tool to think about whatever the challenges that we pose to the tool. There's a lot of value in our unique experience that helps us, ultimately, deliver stronger outputs that are AI-enhanced.
I would like to think that's where we bring a whole lot of strength to this new process that is evolving day-to-day.
“What is your approach to AI tool usage in the face of its constantly changing nature?”
Caitlin
Well, AI is changing every single day and every week and every month, and it's kind of hard to set it and forget it. You can't do that, especially with AI. And so Beth and I, something that we've agreed on, even when we started exploring AI more intensely, is that we want to make sure we're creating a culture of curiosity and clarity. And that's consistent, no matter what the tool or what the process or what the policy is. Sometimes the best thing we can do is just sit with our discomfort and ask, “What problem are we solving for, here?” We say that all the time in our meetings, in our strategies, “How do we make sure that we're solving it the right way regardless of the technology?”
Beth
I love that because it brings us back to the point I brought up initially around making sure we're standardizing everything we can standardize so that we're consistent, both in the way we're working and efficient and with the deliverables we're producing.
One of the approaches that we're taking is, for example, one of our team members, Rachelle. She was involved in one of the earlier video series here.
Rachelle from Day One has been all in on AI, really embracing it and excitedly on the front edge of what AI can do. And I think it's maybe an unofficial title, but she has become our “AI champion” right now. It's not that everybody else is not embracing AI within the agency, but Rachelle in particular is so out there on the front edge of where AI is going that she has been a natural person to do the research and come back to us with ideas around, “Hey, did you know we could be doing this? We could be doing that. I have suggestions for ways to improve this thing we've been doing.” And having one person clearly leading that charge – again, not to diminish the role or the ability of anyone else in the team to do this as well — but it's really helped us because there's such a volume of information out there and it changes every day. It has been a helpful way for us as a small team to be able to, well … I'd like to think that we're ahead of the game. I don't know. I'm humble enough to know that might not be the case. But to at least to keep up with the game, it’s been really, really helpful for us.
Caitlin
I think something that has also been helpful is just to understand that there is a clear driver taking coursework and doing training and being involved in communities and having their finger on the pulse of the bleeding edge of how AI technology is being used. And also just being abreast of all these great tools that have AI built into them. Even from our project management tool, ClickUp, or HubSpot as a HubSpot agency, there's so much AI that's been built into these systems and these products — even integrating ChatGPT into HubSpot and testing and iterating on that. Having a person to bridge the gap between AI developments and our day-to-day operations is something that has been hugely invaluable. And she's been able to develop work groups for specific use cases. So she's not operating in a vacuum, but she's certainly driving.
“There are a lot of AI tools and integrations, and it’s easy to feel inundated by all the options out there. What is your approach to AI tool usage in the face of this constantly changing technology?”
Beth
There are a few layers to the answer. Like Caitlin mentioned, a lot of the tools that we've been using forever now have AI capabilities and functionalities built into the tools. Caitlin mentioned ClickUp, which is the project management system we use. It feels like every day ClickUp is coming out with a new AI tool that you can incorporate.
So there's that level of, “We're already using ClickUp. How can these new functionalities enhance what we're doing to make us more efficient and more effective?” There's that kind of base layer and that feels like a no brainer to us. We've already invested in these tools, right?
The next layer up are things like, “Are we picking a platform such as ChatGPT or Gemini or Claude?” You know, some of those tools, where they're external tools, they're somewhat similar in what they do. There's some overlap in some specific use cases, but how are we making those decisions? Do we have accounts in all three of them? Do we select one particular one?
I think right now we're kind of leaning toward ChatGPT, but there's a lot of evaluation that we have going on, and my answer today on that might change next week or next month. We're trying to make sure that we're leveraging everything in the tools that we've already invested in, in a really smart, deliberate way. And we're trying to be smart and deliberate about the additional investments that are needed.
And that's where it gets a little tricky. What we're doing today might not be what we're doing next month, but we're trying to keep our eyes and ears open and continue to make smart decisions.
Caitlin
I'm going to double down on it's not always about the tool. It's about the process. It's hard to automate or leverage a tool if you don't have some processes in place, but processes need to be teachable. They can't be brittle. We don't have it all figured out, right?
So just figuring out how to build good habits and have some flexibility and lean into our team members. Just like with Rachelle, having AI champions to identify use cases and helpful prompts. We have a Slack channel where people throw in different experiences they have — even by evaluating our shadow self and looking for our shortcomings and being able to level up and be better. It's not always about the tools. And I think sometimes we get caught up in these shiny objects and it can be overwhelming, especially with all the new features rolling out and it can kind of feel like a steamroller, especially for the people who are reluctant to adopt AI.
I think normalizing testing and trying and leveraging AI as a sandbox and allowing different team members — everyone from a junior team member all the way to the CEO — to show and tell, “Here's what we tried either with an operational thing with our project management tool or even a client strategy.” Cultivating playbooks, standard operating procedures, all that stuff, is far superior than investing and testing in shiny objects.
Beth
To piggyback on that, I love one of the core messages you just said: It's about the use case and being deliberate about getting clear on the goal. What are we trying to accomplish with this task or deliverable or project? And then matching the platform or the tool to the use case that helps make sure that we're as intentional as we can be about it. And I love everything about that statement.
Caitlin
I will say that leaning on peer communities has been really helpful. I'm in Charlotte, North Carolina. I often attend the Charlotte HubSpot user group. Charlotte just came out with a new community group called the AI Marketing Exchange. Getting out and shaking hands and talking to people and meeting other sales leaders or agency leaders has helped just kind of normalize the chaos a little bit. But even virtual masterminds like Slack groups … Trust Insights has a Slack community, Analytics for Marketers. Several of our team members are in that and we're able to talk to people all over the world about how they're using AI. Peer communities have been really, really helpful for us.
“What is your best advice to other company leaders who are grappling with a lot of these challenges that you're grappling with right now?”
Beth
Well, it's funny, because I don't know that I'm going to say anything new, but I might pull in some of the comments that Caitlin shared that resonate really deeply with me. First of all, keeping an open and curious mind has been huge for me. There's so much that I don't know. By staying open and curious, I am constantly surprised, like, “That’s amazing – I didn't know I could use AI for that type of project.” I have those moments almost every day, but it's great. That's how you learn.
And then the second is the communities that are available. The Trust Insights community that Caitlin mentioned. The Marketing AI Institute with Paul Roetzer is doing some amazing trainings on this as well. Those are two that I would immediately gravitate to and have helped me.
Again, wrap your arms around what is possible. It’s kind of interesting — as an organizational leader, you need to both have one foot in the future and one foot here, boots on the ground. The peer communities do a great job of helping me understand what is possible and where is this whole world of AI going? What are the best practices that are working for other companies that we can either pick up and implement or tweak and implement? At least understanding how other companies and other organizations are using these tools helps me understand the right way to implement them and document them and make sure the team is comfortable with them. You have to see the forest and the trees, and honestly the communities that I'm in … I'll shout out Reddit. There's some amazing subreddits that bring ideas to the table all the time that are huge and that change the way I think constantly. So yeah, I don't have all the answers. And as a leader, understanding that you don't have all the answers, but having resources to go to for those answers makes all the difference in the world.
“Do you feel that AI has been particularly helpful for you in sort of identifying your own blind spots or are there any instances where AI has eliminated some of those for you?”
Beth
Given my role in the agency as a leader, as the owner of the agency, typically I'm leaning on AI to help me build out strategy. Whether I'm grappling with a particular issue or trying to build a new service line, whatever, I tend to be dealing more with big picture items rather than client work, just because I don't do as much client work, right?
But given that that I'm doing a lot of strategy work and using AI to help refine my thinking and think through problems from multiple angles, one of my favorite things when I think I have a strategy, I put it back into AI. I tend to use ChatGPT. So I'll put it back into ChatGPT and ask it, “What have I missed? Where are my blind spots on this? What gaps does my plan not address? Or what risks have I not thought through? How might this plan go wrong?” That has really helped me take off some of those blinders, right? And see my plan from an even bigger perspective. And once I've identified these risks, now I can build a plan to mitigate the risk.
I mean, no plan is ever foolproof, but that has been game changing for me. I'm definitely someone that I want to charge ahead and move forward with action. And I tend to not stop and assess those blind spots. And so, leaning on AI has really leveled up my own ability to see, to identify holes, so that I can plug them.
Caitlin, what about you? What’s your advice to company leaders who are grappling with this challenge and may feel overwhelmed?
Caitlin
It’s okay to feel overwhelmed. 'Tis life. We are overwhelmed in a lot of ways. There's a lot of things contributing to overwhelm and being suspicious or unsure, but starting small, starting with intention, like literally do not wait for the perfect playbook. If you try to wait for the perfect playbook, it's never gonna happen, right?
So start with things that are lower stakes, lower risks.
Some examples of ways that I've been using AI is within our project management tool and summarizing comments within a certain task.
Another way I've used it is consolidating insights for performance reviews, whether it's team member feedback or client feedback. We do company high fives and shout outs, so capturing those for a team member and analyzing the blind spots that maybe I'm not considering for areas of opportunity or growth opportunities — or even suggested advice and actions that team members can take to grapple with those opportunities.
And even taking that a step further, you could literally put into ChatGPT a strategy or use case and ask it to come up with a ClickUp template for you and you can import it into ClickUp.
These are low stakes sandbox ways that you as a company leader can try things out and not feel like you're going to break them. We were even testing AI agents within ClickUp this week to help us deliver automated reports or agendas for meetings. Again, these are simple ways that leaders can use AI in a way that is not breaching security or going to potentially create unnecessary repercussions.
The big thing is to listen to your people, open the door to re-imagining what you've always been doing. Our goal is not to control AI, but to create a culture where people feel confident and supported using it.
Final Thoughts: AI Needs People
As marketers, the tech's always going to keep changing. That means we have to be agile. What needs to stay steady is not only your trust in your team, but your clients also trusting in you and feeling that you are transparent with them.
As we’ve seen from this three-part video series, AI is only as good as the people behind it. From overcoming hesitation, to finding a healthy balance, to building thoughtful policies, we're all still learning. So, give yourself some grace. And thank you for being a part of this conversation. We're excited to keep learning together.
