Most DTC Discounts Destroy Margin You Didn't Need to Give Away

01 /The Concept

The default marketing playbook for direct-to-consumer brands is discounts and coupons – next-purchase offers, category promotions, limited-time deals. Discounts bring in buyers, but they also eat into margin. The underlying question most brands never rigorously answer: would we have made more money without offering the discount at all? Fewer buyers, sure – but at full price.

Monocle built an AI engine to answer exactly that question. It models future sales both with and without a given discount, compares the scenarios, and identifies which approach generates more profit. The goal isn't to argue against promotions – it's the opposite. The platform finds the specific discount structure that would actually add profit, rather than just add volume at the cost of margin.

Beyond campaign-level analysis, Monocle can personalize offers at the individual customer level. For each contact in the store's database, the AI draws on that customer's purchase history and past responses to promotions to determine: which product categories they're most likely to want, and what's the minimum discount that would actually move them to buy.

All of this runs fast enough to handle millions of customer interactions automatically. And because the platform integrates with popular email and SMS tools, it can regularly send personalized offers across an entire customer base with minimal manual effort.

The workflow is simple: set parameters (offer types, discount bounds), and the AI handles the rest – selecting the optimal offer for each individual customer within those guardrails. Results appear in a dedicated analytics dashboard, where campaigns can be reviewed and parameters adjusted.

The numbers from early customers are strong: revenue per user up 20–46%, cart recovery rate up 30–40%.

Monocle launched its platform last year, closed a small initial round at the time, and has now raised a new $7.5M round.

02 /Why It Matters

Monocle's target market is D2C brands – which typically run heavy promotional calendars – and the agencies that serve them. In one interview, the founders estimate that D2C brands spend around $400 billion per year on promotions, encompassing both media spend and the value of the discounts themselves. If that estimate is in the right ballpark, it's a genuinely enormous market, and investor interest makes complete sense.

The real key play, though, is individual-level personalization of discount offers. Without AI, doing this at any meaningful scale is simply not feasible. The ability to tailor the right offer to each customer in a large database is something that has only recently become tractable – which makes building in this space timely.

OfferFit, [covered previously](/review/kak-povysit-jeffektivnost-reklamnyh-rassylok), is working in adjacent territory. Rather than focusing on the profit math of discounts specifically, it optimizes every dimension of outreach for each individual recipient – including the content of the message and the optimal time to send it for maximum open rate. OfferFit has raised $39M.

Lancey, a Y Combinator alum [covered earlier](/review/na-kazhdom-zarabotat), takes a product angle – using AI to run micro-experiments targeted at automatically identified user segments.

Subsets, another Y Combinator alum [covered in January](/review/privychka-vazhnee-chem-polza), focuses on subscriber retention for digital publications. Its observation: different readers form different habits – some read every morning, some weekly. The platform segments them accordingly and nudges each group in ways calibrated to reinforce their own patterns.

03 /Opportunities

Marketing tactics themselves aren't new. There's very little to invent at the tactic level.

What is new is the ability to deploy any given tactic at the individual level, automatically and at scale. The gain comes not from a better tactic, but from better targeting – and AI is what makes that targeting feasible. The aggregate lift can be significant even when every individual tactic is familiar.

The direction, then, is AI-powered individualized marketing platforms. The most effective ones will likely be purpose-built for specific contexts – like Monocle for coupon and discount campaigns in D2C, or Subsets for subscriber engagement in digital media.

Building a successful startup doesn't require inventing something new. It can mean dramatically improving something old with new technology – the way Uber didn't invent the taxi but made it summonable in five minutes with a phone tap.

The same kind of step change is available in marketing right now, by moving from audience-level and segment-level targeting to true individual-level personalization. The window to build in that space is open.

The most tractable entry point is a vertical where companies already spend heavily on direct marketing – email, SMS, push – and where the underlying purchase data is rich enough to train an individual-level model. D2C e-commerce, as Monocle proves, fits both criteria. The jump from there to adjacent verticals – insurance, telecom, financial services – is a matter of data integration, not a different product.