Hacking GTM Strategies for AI Products with Maja Voje

Can you sell AI products without mentioning AI? 

This is not as challenging as it seems. Just show what the product can do for customers. “The days of selling products just because they are AI-based are over,” believes Maja Voje, go-to-market strategist and founder of Growth Lab. 

No longer does being a first with AI integration lend a competitive edge in a market. You must be different and generate adoption quickly for your product. Even channels hosting users for AI products are not the same. “This is because users are early adopters,” explains Maja. 

In a chat with me, she explains what defines a winning GTM strategy for AI products and urges companies to focus on an early customer profile, not just on an ideal customer profile (ICP). Maja also points to the widening knowledge asymmetry between sellers and users of AI products. What can bridge this gap? Let’s learn. 

This interview is part of G2’s Professional Spotlight series. For more content like this, subscribe to G2 Tea, a newsletter with SaaS-y news and entertainment. 

Sidharth Yadav: You work with companies selling products. How does the GTM strategy for AI products differ from the others? 

Maja Voje: Currently, 70% of my portfolio covers AI-enabled products. I’ve studied this field extensively, and after seeing companies make the same mistakes, there are four things that almost all companies missed on the AI-product journey: 

1. Double down on your ICP and prove it empirically. People ask if a solution is horizontal or vertical. My answer is to launch your product and see who the retained users are. 

It’s nearly impossible to grasp who the ideal user and customer are without seeing some data and without reverse engineering actual usage and monetization data. So, it’s important to empirically check your ICP. Don’t make things up when it comes to personas. 

2. Focus on positioning and messaging. The days of selling products just because they are AI-based are over. We have to find the value propositions for our customers.

This is especially important for new players in the market because you often have to create 10x value to at least have a fighting chance. If you are another AI-enabled sales tool, it won’t cut for you. 

3. When it comes to pricing, we see two extremes. The first one is dirt cheap products that don’t even incorporate the usage cost of an AI model and don’t budget correctly for their costs. 

While users might passionately use these products, the product-maker might lose money in the long run.

The second, which concerns horizontal tools that can serve multiple use cases, is to severely underprice them.  

Please consider raising the prices for such products. Otherwise, it’s difficult to build sustainable business models. 

4. Regarding how to attract users and customers, I prefer inexpensive ways. Why? Because you learn a ton this way. It's very easy to pour thousands of dollars into ads, hoping for the best and reverse engineer data. 

Sign up for G2 Tea.

Marketing news brewed fresh every week just for you. Subscribe here

But when you’re in an environment of unknowns, still figuring out the business models and messaging, even outbound reach would work much better. 

In fact, communities are great, too. Whenever I'm trying to solve a problem in my professional life, I gravitate towards communities. 

“How do you find channels where your customers are? Just ask them where they go to get ideas to solve problems and which people, communities, and influencers they interact with.”

Maja Voje
Founder, Growth Lab

Even established firms don’t present a unique value proposition for AI products. It’s good they have AI. But how does it solve a customer problem? 

I’ve seen many customers going after AI products for the sake of it without being aware of their use cases…

Whenever you have horizontal AI solutions such as customer chatbots or copywriting and video tools, you wonder if they can serve multiple verticals. And each vertical will have a different use case.

Users may lack knowledge about AI, which can create a knowledge asymmetry between sellers and users. The best way to bridge this gap is through use cases. 

People may not be imaginative when using AI. Companies have to present use cases and case studies and sometimes even walk them through demos. With complex technologies, you may even have to engineer a use case. 

You can’t just sit and wait for the customer to understand the use of AI products. It’s essential to give people an idea of how your product can make their lives better. 

What would be the best-performing motions for an AI product in your experience?

First, let’s define GTM motions. Imagine them as a structured way to acquire customers. They are not channels per se. These are more like groups of channels that provenly work well together. I define seven such motions: inbound, outbound, paid digital, community, partners, account-based marketing, and product-led growth. 

There are different parameters to select a motion. First is the cost. Here, companies must pick cost-effective and scalable motions. 

Next, it depends on how customers like to buy. You have to follow their patterns and user journeys.

“I believe you can get to 1,000 users, 100 weekly active users, and 10 paying customers without spending a dime by using GTM motions effectively.”

Maja Voje
Founder, Growth Lab

Today, communities have a big contact potential, especially for AI products. Why? Users of AI solutions are still early adopters, as we would define them along the innovation-adoption curve. Early adopters usually hang around in communities. 

These communities could be Reddit or Facebook groups or even LinkedIn and personal profiles on social media. 

You've spoken about something called an early customer profile (ECP). How is that different from ICP? How can companies successfully chart their journey from ICP to ECP? 

I stumbled upon this concept while working with technical teams, which often would like to avoid human touch.

They would select an ICP based on what their competitors are doing or simply rely on a consultant and brainstorm with them. 

But, usually, this turns into daydreaming about some sort of conglomerate of weird demographic characteristics that don't make sense in a B2B purchasing cycle. 

For example, you may envision your target audience as Hans from Germany, who drives a Volkswagen and loves to ski. This is irrelevant if you are selling accounting software in most cases. 

The only way how to do personas correctly is to reverse engineer data. 

There’s another problem — now versus the future aspiration. You can dream big of working with big brands like Ferrari one day, but without previous references of successful work in the automobile industry, that’s difficult. 

To build your way up, you must have early implementations for good proof of concept. Create case studies for this. You must also distinguish between customers who adopt your solution quickly and those who wait for proof of others' successful use before placing their bets. 

Here, I have to stress the difference between users and customers. You could be a product-led company and have people using your products for free. If you optimize your product roadmap for people who don't pay for it, you could be making a bunch of wrong product decisions. 

Reverse engineer your existing customers, not people who use your product for free and might switch to something else later. 

What do companies get wrong about GTM approaches for newer markets? 

Partnerships definitely work. Local partners can help you shape GTM strategies for a new market segment and help you close deals. 

For example, right now, I'm trying the Spanish market, where I have to price services on the lower end of the spectrum. But, purchasing behavior is different in Dubai. When my Dubai partner saw my prices, they weren’t convinced. They claimed lower prices in that market would imply low-quality products. 

This just means that when you research a new market, don’t compare yourself to billion-dollar companies that have been around for 20 years. If you are a new market player, you will most likely have to play within the 20 to 30% wiggle room of established competitive alternatives. 

Please double down on research and partner up with people with relevant market insights, as you could lose out because of something really petty. It could be as simple as missing a certain trust batch for security or compliance. You won’t know that unless you’re in touch with your target market. 

Many B2B enterprises go into the ‘set it and forget it’ mode regarding pricing. What do they lose out on by not keeping their pricing dynamic? How can they confidently adjust prices to derive real value out of their products?

Pricing can be challenging to manage. It’s the same as posting on Linkedin, which you know you must do, but still don’t. 

It’s essentially the fear of rejection. It’s wired into us. In the case of pricing, it’s loss aversion and the tendency to avoid taking risks for fear of losing clients. 

There is also the conception of fairness. Historic clients could have supported you from the beginning. Having a conversation with them about increased prices can be awkward. 

“Keeping prices dynamic is important for business success. Best-in-class companies change their pricing at least once a quarter.”

Maja Voje
Founder, Growth Lab

It is important to establish the notion of where you’re going. 

Sometimes, with AI, we might even reduce prices. The company that did this very well was PostHog. They said they had optimized processes with AI and would transfer the savings to clients. That’s the news that will never backfire. 

But if you have to raise prices, it gets complicated. The safest way is to reach a new target audience with a new type of pricing. It’s also easier to have custom pricing than a public pricing page. 

If you’re changing prices on your pages, you need to communicate this properly. That starts with transparency, doing it way in advance, and providing people with a good line of support in case they have questions. 

There’s no logic to change prices daily. There has to be some reasoning. I mean, pricing is how we make money. All the other marketing elements spend it. 

When changing prices, I would start with more minor tweaks. This is what I did for my sponsorship pacts.

I always say if the deck is converting at more than 80% then I'm being too cheap. And when I saw the conversion rates from a deck touching 85%, I realized we must bump up the prices. We did it in three different stages instead of one big bump. I recently sold it to two new batches of clients. 

What are some obsolete GTM approaches that companies are reluctant to let go of? What alternatives would you suggest for them? 

Many companies still like doing billboards, trade magazines, and events. This blows me away since I am from a digital-first world. 

Last week, I worked with a corporation in the fast-moving consumer goods space. They were talking about retail marketing on television, and I was like, I haven’t watched TV in the past seven years. 

“You are not your target customer. Rely on channels that your target audience actually wants.”

Maja Voje
Founder, Growth Lab

As for my suggestion, my core message here is that there will always be people who disregard an approach, but outreach will be stronger than ever in 2025. You can make it work by personalizing messages and sending them across multiple touchpoints. There’s even a term for it — GTM engineering.

Things are different in this space than they were 10 years ago. It’s difficult to attribute to a lead now because of privacy concerns. 

Now, it’s about observing the outcomes of your actions. If you’re organizing events, observe them for at least one to three months to get relevant data. 

The message is clear: do what works best for you. If your target customer buys newspapers, then you should be in newspapers.

Many companies face the pressure of being first-movers in a market. Is that still an advantage? How can late entrants make their GTM strategy work? 

The Blue Ocean Strategy was popular 10-15 years ago. 

But today, the market is saturated with AI solutions. It's no longer about being the first one out there but about being different and generating a critical mass of adoption as soon as possible. It doesn’t matter who’s the first in the market. 

For newer players, I’d try to locate a friction market that is underserved by big competitors in a space. 

What are some of the low-cost, high-return strategies for small companies?

There are three popular ones in my community: 

1. Close the first million dollars in sales with a single case study. This is a very bold hypothesis of a radical focus. If you do something well for one client, send the case study via outbound to people just like them. Solve the same problem repeatedly and build a credible track record. 

2. I call the second strategy social programming. 

It is about leadership: leading with visions, unique charisma, and personality. This is how you can win against all odds. You could be radically different, authentic, and sometimes controversial too. 

It wasn’t easy when I launched my book. I am from a small country called Slovenia, didn’t come from an Ivy League school, and the odds were against me. 

To cut through the noise, I had to build a demand generation machine on LinkedIn and Substack. The only way to do this is to have proprietary (thought leadership) ideas and follow a unique path. 

3. Automate your data generation. What we’re seeing in terms of outreach and data enrichment is interesting. So, use a tools stack to personalize messages and create a sales machine. 

What's the future of GTM motions, and what trends will shape them in 2025?

There are trends and anti-trends.

On one hand, there are AI enrichment tools that make it easier than ever to target a customer and automate processes. 

On the other, there’s a focus on hyper personalization. People sell services through personal messages on LinkedIn and hold weekly calls with their community members. 

So it depends on which side of the spectrum you want to be on. While some channels might die off or slow down, there will always be alternatives.

In today’s digital-first world, we must build genuine relationships of trust based on human-to-human connection. This is also important to sales, where empathy is important. 

Empathy is the antidote to AI. Both empathy and the use of AI are required in combination, of course, so learn them. 

Because of increased pressure to show a return on investment (ROI), companies are slated to drive 50% more utilization of purchased AI products. Learn about gen AI trends of 2025 here.