Where Should Brands Start With AI

Everyone is trying to figure out their strategy for how to adapt their business to the age of AI. As a freelancer, I am frequently asked what I can help brands do. It's a reasonable question when you consider that most freelancers have a specialty; something like media buying on Meta, CRM on Klaviyo, or managing the storefront on Shopify.

I spent the last fifteen years building software as a product manager, with the last four years specifically working in the e-commerce space. The reason I have a hard time answering this question succinctly is - what used to take an entire development team and months of work can now be done just using Claude in a matter of days. Because of my experience building software and also working inside of brands, I'm an AI Swiss Army knife. 

When it comes to AI and how to implement it best into a business, there are a few factors of how opportunities should be identified and prioritized. 

What are the Current Company Dynamics?

Where is the company in its growth trajectory? Is it founder-led or professionally managed? Has the company taken capital from venture or private equity? All of these dynamics play a big role.

For example, if the company is relatively young and at the early part of its growth curve, I feel there is a huge opportunity to tackle each new chapter of the business from an AI-first perspective. Investing in process and technology and data structures early on makes it possible to keep up with the rapid pace of AI development rather than having to redo things that have been in place for years. Brands that get this right have the opportunity to continue growing and adding new sales channels or new marketing channels or new supply chains, etc., without adding any incremental headcount.

Brands that are farther along and already have been built out with a team of specialists have a totally different task at hand. They have to figure out which parts of their business are well-suited for being adapted with AI and where it will make the biggest impact on the bottom line. They also have to factor in the specific talent that they already have on hand and whether or not they are a good fit for the way work is evolving.

In-House or Agency. When Does Each Make Sense?

This brings me to the perennial question of whether or not to do things in-house or to work with an agency. Agencies can provide a lot of value because of deep expertise and what they see across their pool of clients. They may also have the ability to call on top-tier talent who are highly specialized.

Typically, brands bring on agencies to keep their cost structure in line. You can bring on multiple job functions for a monthly retainer far less than what it would cost to hire multiple full-time people.

I think the narrative is changing here, though, for brands that successfully adopt AI tools. From my own experience, it's absolutely possible to have one person who manages creative strategy and insights, media buying, as well as influencers and UGC. This person needs to have expertise across all of those domains as well as be strong at using AI tools.

Candidly, when a brand is working with an agency, I tell them there isn't much that I can help them with in that part of the business, because if the agency is successfully adopting AI, it improves their operating margins, but they're not going to pass that savings on to the brand.

Improving Operational Execution and/or Getting to Profitability

E-commerce has become an increasingly challenging space over the last few years, with sky-high acquisition costs and tariffs driving up the cost of goods, not to mention a softening macro economy from the consumer perspective. As a result, a lot of brands are trying to figure out how to get to or maintain profitability, let alone building the P&L that they would like to see.

This is where selecting AI priorities really becomes cut and dry, just by looking at the P&L. The first places I would look with any brand are the following:

• Customer support costs
• Revenue per head
• Content costs
• Software contracts
• Primary e-commerce funnel metrics

All of these are addressable with the right AI strategy.

Adoption

‍

It's an adoption problem. It’s a strategy problem. It's an executive education problem.

‍

When I hear anecdotes or see headlines about companies using AI but not finding any short-term ROI, I chalk it up to poor prioritization. I am 100% sure it's not a technology problem in most cases, because I am seeing the benefits myself.

It's an adoption problem. It’s a strategy problem. It's an executive education problem.

I would be the first to applaud any brand that is taking AI education seriously for its employees. Providing them with resources and ongoing support to make sure that they are able to find ways to adopt AI is a great investment.

However, giving everyone on your staff an AI co-pilot and hooking them up with some AI training is a great first step, but it will not be productive for 80% of people. 

I have a lot of experience building software for process automation, both in e-commerce and within SaaS companies. And unfortunately, I'm here to tell you that not everyone has strong systems thinking, analytical skills, or a knack for building processes that scale.

The critical exercise that leadership at brands need to undertake is mapping out their business processes, tying them to specific individuals at the company who have the most organizational and domain expertise. Understanding which parts of their business operations can be automated or supported with better analytics, and ultimately rebuilding their teams/processes to be complementary to AI.

I'll give you an example to help you understand why AI really only works if it's transformational. Imagine you're running a business before the advent of personal computers. You start to hear about businesses transforming the way they operate and cutting costs by relying on this new digital technology. You order your first couple pieces and give them to critical staff. Everyone has a confusing new sidekick they have to learn to use on top of all of the other stuff they have to get done for business as usual. The company hasn't really transformed, and everything is still done the same way, except now critical people have a distraction they're trying to figure out how to use.

Compare that to the way companies operate today. Everyone has a personal laptop for their own work. They bring it home with them at the end of the day. They literally cannot do their job without it.

It will take time for this transition to play out, but I think the example is clear. Only meaningful transformation will prove the value of AI.

‍