It’s been bothering me for years. When you read the rosy hype about AI in the new business environment, you could be forgiven for thinking it’s easy.
It isn’t, and it can’t be.
Every business is essentially a custom design environment. You have to work with a range of products, multiple market and commerce considerations, and much more.
AI has actually made business far more complex. It’s not just AI, either, by any stretch. It’s a whole added dimension. The entire configuration of your business hinges on a smooth, trustworthy commercial model.
You have to integrate your technologies, create a business strategy and probably a whole new business model before you can do business at all. Marketing alone can absorb a lot of time, money, and effort. Getting any product online is tough. This isn’t minutiae; it’s core business from inception.
The considerations equate to “get it right from day one” as the real criteria. The legendary massive AI failures alone are a clue as to how difficult this work can be.
I’ve been looking at this mess for a while. I was wondering how people were approaching the new horizons.
To explore what adaptation looks like in real commercial environments, we spoke with Sofia Toebak Bravo, an international growth strategist whose portfolio spans luxury fashion, apparel and accessories, beauty, consumer goods, B2B SaaS, and premium DTC brands. She has led multi-market growth programs across North America and Europe for the better part of a decade, managing channels that generate seven-figure annual revenue through affiliate and partnerships ecosystems.
She is best known for designing scalable acquisition frameworks for net-new programs and aligning cross-channel measurement to reduce wasted spend. She has worked with PartnerCentric, a New York-based agency recognized for supporting affiliate expansion for leading brands.
OK, say you’re a business trying to do business.
These questions are unavoidable for businesses:
How do you even get started?
You can’t figure out the ‘how’ before you figure out the ‘what’ – so that is your first step. What is success to you in this scenario? Work backwards from there.
Most businesses begin with a product idea and a marketing plan, but in the AI era the first real question is: what data, systems, and workflows will actually support your goals? It’s so much more efficient to build a scalable foundation than adapting and revamping and redefining continuously as your starter systems quickly become obsolete or can’t sustain your goals.
Look for your tech stack, what customer acquisition channels are going to be your growth levers and how to optimize them. Then map out how AI can remove or reduce friction, add ‘bulk’ to your organizational power and fuel your ramp-up.
How do you plan?
When planning net-new strategies, I always build in layers: business model first, channel strategy second, and execution plan third. AI fits into that as a force multiplier, but only if you’ve already defined what you’re optimizing for: padded margins, rapid ramp and exponential growth, customer retention and life cycle, or market penetration and international expansion.
Depending on your choices, different channels will be your best friend. If looking to penetrate a brand new market, effective educational creative paired with robust partnerships and paid media will yield the fastest result with strong reputation outcomes, whereas if pushing for customer retention and life cycle you want to add the most focus on customer service, education, and CRM.
What are the risks?
The biggest risk is businesses deploying AI before they have enough training internally and clarity on how and what they’re implementing for.
In my work, I’ve seen three common failure points:
- poor data quality and measurement, fragmented or siloed reporting and ignoring interactions across teams and channels is always an easy mistake to make.
- fragmented tools and teams, feeds into my first point but you’d be surprised how many teams just don’t talk to each other!
- Tools and processes that reward speed over accuracy
In affiliate and partnerships specifically, AI can accelerate partner discovery and optimization, but it can also accelerate bad decisions if attribution, compliance, and brand safety aren’t built into the model.
Can you model your options to scale?
Yes, but not with one spreadsheet anymore. Scaling requires scenario modeling that includes channel mix, partner economics, supply constraints, and cost volatility.
I like to think about it as building multiple “futures” not one forecast. You want to know what happens if: paid media costs jump up, a partner channel outperforms or underperforms severely against expectations, or conversion rates change due to creative iteration. Basically prepare for all options and you’ll have the best chance at success.
AI makes this much much easier, but only if the inputs are clean and the model is grounded in real performance data, this is why team knowledge sharing and education are core to true AI native companies, talking to AI is not enough if we don’t know how to do it effectively.
What level of comprehension do businesses have of this new environment?
It varies dramatically. Most executives now understand AI conceptually, but I’ve not encountered many that understand how to practically interact with it in a native way, integrating AI into decision-making, measurement, and execution without losing control of risk, brand, or costs.
A common pitfall since the AI ‘boom’ has been companies wanting to claim AI literacy without doing the internal work to make this a reality beyond tool subscriptions and accessibility.
What level of control can you have over your costs?
More than most people think! But only if you build the right system.
AI can drastically reduce costs in creative production (a huge win for most marketing channels), forecasting and financial modelling and customer service support. But AI can also create hidden costs: tool sprawl, duplicate work, compliance exposure, and inefficient testing.
Pay attention to where AI is allowed, where it’s supervised, and how performance is measured, and above all equipping the right people with it rather than blanket basic use.
Can you create an integrated marketing strategy and business model from scratch?
Marketing is part of the business model now, whereas it used to be an operational department only. In my work we spend a lot of time educating ourselves, brands and partners on cross channel interaction nuance. It’s key to have an integrated strategy that connects product positioning, channel economics, acquisition and partner strategies, retention, and success measurement.
I often see affiliate and partnerships looked over, which I urge new businesses to avoid, because they allow brands to scale customer acquisition with more control over ROI than many paid channels.
What are the risks for new business ventures?
In fear of sounding like a broken record, early risks tend to be one of two things: overconfidence and under-infrastructure.
Many founders assume they can “figure it out as they go,” but today’s market punishes improvisation. The ventures that succeed build measurement and operational systems early. That includes attribution, compliance, and channel diversification.
How do you manage market penetration?
Above all, you need a channel strategy that provides a good balance of speed and credibility.
In my experience, paid media, partnerships and affiliate ecosystems can accelerate penetration because they give brands access to audiences that already exist through publishers, creators, loyalty platforms, and niche communities aligned with their goals.
Working in stages is key and once you realise this it seems almost obvious! establish proof of conversion in a narrow targeted segment, and expand and repeat into adjacent audiences.
Let’s not oversimplify. Many businesses manage critical add-ons like affiliate programs, reseller options, and work in huge marketplaces like Amazon. All of this needs mapping before you can start operations.
That’s where you start, folks.
