Building a Marketing Function in the Age of AI

In marketing, one of the buttons we commonly push is the “more” button. Not getting enough leads? Let’s produce more content. Need to increase the metrics on a campaign? Let’s spend more on it. 

It’s tired. Frankly, it’s not always the right move when what companies need isn’t more; it’s focus. 

How Has AI Changed Marketing? 

With the rapid development of AI tools, you’d think building and auditing marketing functions should get easier for businesses.

The abundance of AI copywriting, content-creation, and marketing-automation tools should make it easier for small teams to build highly impactful marketing functions without a ton of overhead, and in some cases, that’s true. 

In my experience, the more prevalent issue is that there is so much information and so many different tools with varying capabilities that teams don’t know where to start. 

You can consult your AI companions, but because AI has the knowledge of the whole internet at its disposal, and because it’s completely reliant on its available knowledge and your prompting skills, it’s likely to lead you down the wrong path at least a time or two. 

When it comes to building or rebuilding an impressive marketing program in the modern age, my two cents are: 

1. Pause to look inward, and 

2. Less is more. 

   
   

With so many tools at our disposal, it’s tempting to jump right in, adding new programs, software, and AI tools without stepping back to ask whether they’re necessary or how they affect customers.

In my experience, however, it’s essential to identify the pillars of your brand and marketing strategy before adding a bunch of software and tools. The philosophy is simple: it’s much easier to boil a pot of water if you don’t continuously add more water.

Philosophy #1: Pause to look inward 

Whether you’re building the function from scratch or trying to focus your marketing efforts, start by taking a long, strategic look inward before adding more to your plate. Do a thorough investigation into your company, solutions, branding, positioning, and team: 

Ask questions of everyone in your company. 

Whether you’re brand new or you’ve been there longer than your peers, you want to get a good grasp on how your team perceives your organization. 

Pretend like you know nothing — like it’s your first day, and ask everyone what they understand the organization's mission and goals to be, and how they explain what the company does to other people. 

Ask them to explain their role in the organization and how they contribute to its long-term success. Ask about who you serve, how you serve them, and how good you are at it, especially compared to competitors. 

Question their perception of the org’s positioning, key segments, customers, and competitors. Conversations about ICPS and Buyer Personas are incredibly enlightening, especially when it comes to uncovering pain points and motivations. 

Collect data from these conversations and analyze it for consistencies or misalignment. If your team each explains your organization and services differently, you’ve probably got a messaging or internal alignment problem.

You can uncover truths that employees may know, but the organization has never named aloud. 

By asking a million questions to everyone at Idle Smart and sitting on ample customer calls, I quickly learned that no one on our team described the organization or what it does the same way. There was no differentiation between the Idle Smart system and Idle Smart the company, which created confusion internally, let alone for customers. The positioning was also unclear; as a category creator, the organization struggled to understand where they fell in the market and how competitive they were. There are a million other insights I found that I could share, which is exactly the point of this process. 

Because of my research, I had enough information to create our original Mission, Vision, Values, and positioning. Plus, the organization not only settled on a common understanding of what we do and how we do it, but also created a distinction between the company and product, and the brand to its competitors in the market. 

Especially as AI creates more content, it’s increasingly important to ensure your messaging aligns with your customers’ actual needs and behaviors, not just with what your team thinks customers need. 

Do objective research. 

Pretend you don’t know anything about your company. Research it as if you were a very critical buyer looking for the best solution for the best price. 

Search the website and socials top to bottom, looking for answers to specific questions your customers would ask about pains they experience every day. 

• Do your marketing materials answer your questions? 

• How easy is it to get answers to your questions? 

• Is it easy to use and pleasing to the eye? 

  
  

Research competitors in the same way and “make a purchase decision.

• Who would you purchase from and why? 

If it’s you, and you’re truly, objectively best in the space, then great! Dig deeper. 

• How close of a competition was it, and why? 

• What would have helped you make that decision faster? 

  
  

If it’s your competitor, figure out why customers might choose them over you, and use that to tailor your strategy. If there are many areas for improvement, just tackle it one bite at a time; don’t try to swallow the whole apple. Start by surpassing your competition in just one area, or by answering just one question better than they do. 

Talk to customers (duh). 

Customer research is always full of gold. It’s a little harder for smaller orgs to do because we don’t always have the resources to implement a standardized customer feedback method. Regardless of the company size, it’s always, always, always worth investing in customer research. 

Collect customer feedback on their satisfaction with the product or solution, your team, and the overall customer experience. Data can tell you a lot about how people use your website, but talking to them about their thoughts and feelings, and asking them how you can make things even easier for them, can reveal a lot about their psychology. 

Make sure you’re not asking leading questions and invite detailed answers with open-ended questions. Start with a key research question(s), and if your conversations with customers continuously shift away from answering that question, consider re-evaluating your questions to make sure you’re really digging into the heart of your issue. 

You can start this process informally by slipping questions into conversations you’re already having with customers or asking for feedback via email. You can also use tools like HubSpot or SurveyMonkey to collect data via surveys. 

There are several free tools you can use to get data as well. Look at Reddit, sites like G2, and even Glassdoor to learn what people are saying about your company. Sometimes it’s unpleasant to see people slamming an organization you’re proud to work for, but it’s much easier to get ahead of a potential PR catastrophe when you’re aware it’s a pain point for customers. Plus, getting surprised by a nasty Google review is a lot easier to take when you have a plan of how to respond and are actively working to right the issue.

Your goal at this stage is simply to be a sponge; just absorb the information and use it to understand where there are marketing gaps. Once you have collected enough data to make strong conclusions or suggestions, you can take your findings to leadership or your team with data-backed confidence in your recommended action plan. 

The best part about conducting marketing research in this way is that the most immediate, most painful problems often jump out at you. They tend to come up over and over again.

If you are just getting started and don’t know where to begin, this research will point you in the right direction. Plus, it gives context into who is affected, how severely, and the priority level of the issue (and resources likely to be lent toward solving it) across the org.

Don’t get caught up on using the right software or even the lack of quantitative data, especially if you’re at a startup without a ton of resources or capital. Sometimes the only data you’ll get is qualitative, and let me tell you, that’s a million times better than no data at all. The worst thing you can do is not collect any data because it’s not quantitative enough. 

Leadership may want quantitative data (don’t we all) for reporting and ROI analysis, but don’t be discouraged; it’ll come. The qualitative data you collect may not look flashy on a board slide, but that’s where you find pain points Google Analytics can’t capture, giving you an opportunity to delight and raise your customer LTV (which will look good on a slideshow). 

Philosophy #2: Less is more

This is a particularly hard principle to follow these days, given the millions of tools available at our fingertips, but the fewer tools you use, the smoother your marketing ops will be, especially when you’re just building your marketing department.

Think about it: on your first day at a new company, would you rather log into 100 different systems and spend 90 days just trying to wrap your head around them? Or would you rather get 10 systems, spend 30 days learning them deeply, and 60 days learning about the company and figuring out how to optimize them? Obviously, there’s no all-in-one solution for every single one of your business needs, but having a unique solution for each problem creates friction, even when it’s advertised to reduce it. 

Choose your systems carefully. Look for technology that can solve current problems and that won’t become a hassle in the future. 

For example, if you choose a web hosting platform and later find out it doesn’t help maintain GDPR privacy standards, which are not enforced in the U.S. yet but are likely coming, you may soon have to purchase an additional plugin, adding another solution on top of your site just to maintain legal compliance. 

Some solutions are much more integration-friendly than others; when systems are integrated, you can speed up your processes and automations immensely. Plus, smooth integrations reduce the friction between systems, making your tech stack significantly easier to use. 

Technology is helpful, to an extent. When we introduce technology where it doesn’t need to be, we over-engineer, overlook optimization opportunities, and make our work more challenging. 

I’m not suggesting technology avoidance. I recommend being very strategic with your tech stack. Instead of finding 12 solutions that each solve a problem perfectly, I challenge you to find 2–3 key systems to build your foundation on, with the expectation that they may not solve every problem. Where your most critical, highest-impact gaps are after implementing your core tech stack is where you can start to think about adding specialized solutions. 

Auditing existing marketing programs

Building a marketing program isn’t exclusive to startups; this is also an incredible practice for companies with some kind of marketing in place.

Especially if you’re heavily using AI, it’s a critical practice to ensure your messaging doesn’t stray from reality. Trends are moving faster than ever, so realigning marketing strategies and continuously conducting customer research are increasingly important for brands to ensure they remain aligned with their customers and target markets.

When I talk about this approach for orgs with existing marketing programs, I like to think about it as “stripping marketing down to the bones.” It’s my favorite method for reaching the heart of any company and identifying gaps between who you are in the eyes of your customers and your employees. 

The process is essentially the same as for startups, but established companies have way more resources and easier access to customer data. The problem is not “what to add,” but “what can we cut?” 

Your goal is to peel away all the layers that make up your marketing program and understand why they are there, who they help, and find opportunities to streamline operations, systems, and strengthen your brand. 

The kicker is that they also have an established brand that can be hard to audit when they’re very well known. Pretending like you know nothing about the company is even more important at this stage, but it’s a million times harder. 

Interviewing your team is still important, but it’s usually not possible for large orgs. If you have a big team, rather than trying to talk to hundreds of people, select a subset of that total who you want to talk to. Get a few people from each team with various years of experience, or even consider having these conversations in groups to get feedback from multiple people at a time. You can determine how many people you want to talk to and for how long, but when you look back at your interviews, you should see a fairly even, random spread of titles across the organization. If you have a heavy concentration of leaders giving their opinions, you should probably speak to as many, if not more, people who are not in senior leadership. 

If you have a standout intern, definitely talk to them. Getting feedback and ideas from people who don’t know the company well, and may not even have another reference for how businesses should operate, can open doors you may have never considered. Interns often don’t get to know the ins and outs of a company quite the same as a full-time employee, so they know less about the limitations and history. Without those constraints, they can provide incredibly helpful insights and brilliant ideas. Their suggestions may not be realistic, but there’s almost always a way to work around constraints, even if that’s reimagining the idea. 

Moral of the Story: Build the Tracks Before Adding New Cars to the Train

Take the time to learn yourself and your brand, and your positioning will fall into place, and better yet, it’ll be backed by data. 

Restraint is a competitive advantage. 

Especially in the age of AI. 

By pausing, looking inward, and resisting the urge to add more before the marketing bones are sturdy, you’re more likely to build a marketing function that works, scales, and excels. 

Once you have the foundation built, then you can start stacking bunches of fancy tools.