# Depict.ai Financial Model

AI-powered product recommendation engine delivering Amazon-quality personalisation to any e-commerce store [DECK slide 1]

- Canonical: https://finamodel.com/startups/depictai
- Excel download: https://finamodel.com/startup-models/depictai.xlsx
- Category: AI/ML
- Model type: SaaS ARR / Valuation
- Funding round: Seed
- Funding: $17M
- Founded: 2020
- Geography: Sweden / Nordics (customer logos are Swedish brands; Staples Europe expansion mentioned) [DECK slides 3, 4]
- Customer: B2C

## About the company

Depict.ai provides product recommendations for e-commerce stores using both images and text, rather than relying solely on transaction history. Its multimodal approach is designed to help smaller merchants deliver Amazon-quality personalisation even when they lack the behavioural data of a large retailer.

The company runs a free two-week A/B test, then takes 10% of the incremental revenue it generates. The deck assumes a 4% merchant-GMV uplift, making revenue scale with each customer's GMV; it lists 21-plus live customers and an expected Staples Europe expansion.

The model builds monthly merchant cohorts and applies GMV, the delivered uplift, and the performance fee to calculate MRR. It tests onboarding pace, same-store GMV growth, churn, Staples timing, infrastructure gross margin, sales and R&D expense, cash burn, and runway.

## What's included

- 5-year monthly revenue build with stage-appropriate growth assumptions
- Full P&L, headcount plan, and operating-expense schedule
- Cash-flow statement, runway, and burn-rate tracking
- Valuation via exit multiple with a DCF cross-check
- Returns analysis with MOIC and IRR
- Unit economics including CAC, LTV, payback, and cohort retention

## Product & value proposition

- Recommendation engine trained on images and text - not solely on transaction data
- Proprietary edge: multimodal understanding (vision + NLP) allows accurate recommendations even for small stores that lack transaction volume
- Demo product live at demo.depict.ai
- Positioning: "Amazon-quality recommendations for any e-commerce store"

## Market

- Global e-commerce market: $3T
- Average revenue uplift Depict.ai delivers to customers: 4% of merchant GMV
- Depict.ai take rate on that uplift: 10%
- Derived TAM: $3T × 4% × 10% = **$12B**

## Revenue model

- **Performance fee**: Depict.ai takes 10% of the incremental revenue uplift it generates for each merchant
- Mechanism: free 2-week A/B test to prove lift → monthly recurring fee = 10% × measured incremental revenue
- Fee is variable (scales with merchant GMV × uplift rate) - not a fixed subscription
- Sales motion: outbound demo showing comparative recommendations → A/B test → conversion to recurring

## Traction & metrics

- Named customers: 11 live (RoyalDesign, KitchenTime, MEDS, Svenskt Tenn, Junkyard, Nudie Jeans, Edblad, LloydsApotek, Dogman, Grandpa, Staples) + "+10 others" = 21+ customers
- Staples signed; contract expansion to all of Europe expected in 2–3 months pending Intershop platform launch

## Unit economics

- A/B test results:
  - vs. AWS Personalize: **2× increase in click-through rate**
  - vs. Nosto (largest competitor): **150% increase in add-to-cart**
  - vs. in-house data scientist (KitchenTime): **270% increase in recommendation revenue**
- Average revenue uplift claimed: 4% of merchant GMV

## Competition / moat

- Named competitors: AWS Personalize, Nosto; in-house data-science teams
- Moat claimed: multimodal AI (image + text understanding) vs. purely transaction-based systems; cross-merchant data network effect implied by slide 6 (individual stores lack data; Depict.ai aggregates across stores)
- Competitive differentiation: 2–6× better recommendation performance claimed

## Team & funding ask / use of funds

- Oliver Edholm, CEO - described as "youngest AI researcher in the world", ex-Klarna AI research, age 18 at time of deck
- Anton Osika - first employee at Sana Labs (30+ employees), ex-CERN
- Institutional affiliations shown: Klarna, NUS (National University of Singapore), Babyshop Group, Sana Labs, CERN

---

## Recommended financial model

- **Archetype + why**: **Performance-revenue SaaS / GMV-linked revenue model** - Depict.ai's fee is a variable % of incremental merchant GMV, not a fixed subscription. The model must flow from (# merchants) × (merchant avg GMV) × (uplift rate %) × (take rate %) = Depict.ai revenue. A monthly-cohort MRR build is appropriate since fees recur monthly and scale with merchant growth.

- **Forecast horizon & granularity**: 3 years monthly (Y1–Y2 monthly detail, Y3 annual summary); monthly is needed to track cohort ramp and merchant onboarding pace.

- **Key drivers & assumptions**:
  - New merchants signed per month
  - Average merchant GMV at onboarding
  - Average revenue uplift rate delivered: 4% of merchant GMV
  - Depict.ai take rate: 10% of uplift
  - Implied Depict.ai revenue per merchant = merchant GMV × 4% × 10% = 0.4% of merchant GMV
  - Merchant churn rate (monthly)
  - Merchant GMV growth (same-store sales growth)
  - Gross margin
  - Headcount plan / opex
  - Staples Europe expansion as a step-change event

- **Scenarios (Base / Bull / Bear - which variables flex)**:
  - **Base**: 3 new merchants/month ramp, average merchant GMV €2M, 1.5% monthly churn, 75% gross margin
  - **Bull**: 6 new merchants/month, Staples Europe lands in month 4 (large GMV step-up), 0.5% churn, 80% GM
  - **Bear**: 1–2 new merchants/month, Staples expansion delayed/cancelled, 3% churn, 65% GM (higher infra costs)
  - Key flex variables: merchant adds per month, average merchant GMV, churn rate, Staples Europe timing

- **Required sheets / outputs**:
  1. `Assumptions` - all inputs with / tags
  2. `Merchant Cohort Build` - monthly new merchants, cumulative active, churn waterfall
  3. `Revenue` - GMV per merchant × uplift × take rate → MRR → ARR
  4. `P&L` - revenue, COGS (infra/hosting), gross profit, opex (S&M, R&D, G&A), EBITDA
  5. `Cash & Runway` - cash balance, monthly burn, months of runway
  6. `Scenarios` - Base / Bull / Bear toggle
  7. `Dashboard` - KPI summary: active merchants, ARR, MRR growth, gross margin, runway

## Frequently asked questions

### Is the Depict.ai financial model free?

Yes. The Depict.ai model is a free Excel download with live formulas.

### Can I change the assumptions?

Yes. The workbook is editable and its live formulas recalculate when assumptions change.