Start with the economics, not the channel
Almost every stalled e-commerce brand we meet is trying to solve a maths problem with a tactics answer. They are testing new platforms, new agencies, and new creative formats when the underlying issue is that the first order does not pay for the customer who placed it, and nothing in the business is designed to make the second order happen.
Before you touch a campaign, write down four numbers: average order value, first-order contribution margin after cost of goods and fulfilment, repeat purchase rate at ninety days, and current blended acquisition cost. Those four numbers tell you whether you have an acquisition problem, a margin problem, or a retention problem. Only one of them is fixed by better ads.
- If contribution margin is thin, fix pricing, bundling, or fulfilment before scaling spend.
- If repeat rate is low, invest in lifecycle and product experience — paid will only accelerate churn.
- If both are healthy and growth is flat, you genuinely do have an acquisition problem worth spending against.
Positioning is a performance lever
In a crowded category, the brands with the lowest acquisition costs are rarely the ones with the cleverest media buyers. They are the ones whose reason to exist is obvious in three seconds. Positioning is not a branding luxury that comes after growth — it is the single biggest determinant of how expensive your clicks are.
The most useful positioning work for a DTC brand answers three questions in plain language: what does this replace, who is it obviously for, and what is the one thing a competitor cannot honestly say. Every strong ad angle you will ever run is a variation on those answers.
Treat creative as the real targeting
Platform targeting has largely commoditised. What decides who sees your ad, and whether they stop, is the creative itself. A founder explaining why the product exists will outperform a polished brand film more often than agencies like to admit.
Build a testing roadmap rather than a content calendar. Each batch should test one variable — a new angle, a new format, a new proof point — so that when something works you know why. Retire fatigued assets on a schedule instead of waiting for performance to fall off a cliff.
- Angles: problem-first, founder story, ingredient or craft, social proof, comparison, objection handling.
- Formats: static, UGC-style video, demo, unboxing, before and after, text-led.
- Cadence: a fixed number of new concepts per month, with iterations on winners weekly.
Build the funnel, not just the top of it
Most brands over-invest in prospecting and under-invest in the two cheapest stages of the funnel: retargeting people who already engaged, and reactivating customers who already bought. Those two audiences are where margin lives.
A working structure separates prospecting, retargeting, and retention so that each can be read and funded independently. When they are blended, a healthy retention programme can hide a broken acquisition engine for months.
Lifecycle marketing is where the profit is
Email and SMS are the highest-margin channels in e-commerce and the most consistently neglected. The flows that matter are unglamorous: welcome, abandoned browse and cart, post-purchase education, replenishment, and win-back. Built properly, they lift lifetime value enough to change what you can afford to pay for a new customer.
Once lifetime value rises, everything upstream loosens. You can outbid competitors, tolerate a longer payback period, and scale channels that looked unprofitable on a first-order basis. This is what compounding actually means in practice.
Measure what you can act on
Attribution will never be perfect. Rather than chasing certainty, pick a small set of numbers you trust and review them on a consistent rhythm: blended acquisition cost against contribution margin, new versus returning revenue split, ninety-day repeat rate, and creative-level performance.
Review weekly at the creative level, monthly at the channel level, and quarterly at the economics level. Changing strategy faster than your data can move is the most common way brands talk themselves out of something that was working.