Pangolin
Login
Welcome back! Please enter your details.
Your P&L doesn't stop at revenue. Neither should your media model.
Pangolin is built around DTC economics specifically COGS, returns, and LTV, not just spend and top-line revenue
Defending margins, not just tracking top-line sales
Most measurement tools, including a lot of enterprise MMM, stop at revenue. For a DTC brand, that's the easy part of the P&L. Returns can run at 20-30% in some categories. COGS varies by product. New customers cost more to acquire than repeat ones convert for. A model that optimises for revenue alone will happily recommend spend that looks good on the top line and quietly erodes margin underneath it.
Continuous Bayesian models that speak DTC economics
Pangolin's model is built around contribution margin from the outset, not revenue with margin bolted on as an afterthought. Where connected, COGS, return rates, and CRM-derived LTV feed directly into the model, so a channel's budget recommendation reflects what it actually contributes to profit.
Built by a team who have built and used MMMs to supercharge growth
DTC needs continuous iteration, not 6-month consulting cycles
Enterprise MMM tends to be built for slower-moving businesses with stable channel mixes. DTC brands change channel mix quickly, launch and kill campaigns fast, and often don't have years of clean history to model against. Pangolin's Bayesian approach is built to produce a usable, honestly-uncertain answer with the data a growing DTC brand actually has, and to keep updating as that data grows.
Rethink growth with margin-level evidence
Pangolin turns complex probabilistic calculations into solid financial guardrails for your e-commerce operations.
Frequently Asked Questions
Clear answers about Bayesian model integrations and data pipelines for e-commerce.
We don't have years of clean historical data - is this still useful?
Yes, the Bayesian approach is designed for exactly this. Priors handle limited history so you get a usable, honestly-uncertain answer from the data you actually have, without needing 3 years of perfect tracking.
Do you need our COGS and returns data to work at all?
No, core MMM runs on spend and revenue. Margin and returns data sharpen the model but aren't required to get started. You can integrate basic Shopify revenues first and hook up COGS streams later.
Is this built for a specific product category?
Pangolin works across DTC categories - from high-return fashion and apparel to high-LTV cosmetics, nutrition, and subscription boxes. The model adapts to your specific unit economics regardless of vertical.
Which ecommerce and CRM platforms do you integrate with?
Pangolin connects directly to Shopify for revenue and order data, with COGS, returns, and LTV feeding in from your connected CRM or ERP system. If a platform isn't natively supported yet, data can be brought in via a standard export while direct integration is added (this usually takes 5 working days).
We sell multiple brands or a wide SKU range with very different margins, can the model handle that?
Yes. Margin can be modelled at category level where SKU-level margins vary widely, so a channel driving a high-margin category and one driving a low-margin category aren't blended into a single misleading average.
Stop guessing. Start proving where the real margin is.
Free consultation for brands spending >500k / month.