# Finsi - Do More With the Team You Have: AI Agents for E-commerce, Under Your Rules
# Full Documentation

> Finsi takes the spreadsheet work and the bottlenecks off e-commerce teams. AI agents work from one set of verified numbers and follow the brand's rules: every agent asks before it acts, leaves a record, and stays inside its access. Brands bring their own agents (Claude, ChatGPT, Codex, Grok and others over MCP or API) or start with Finsi's Agent Library. Finsi connects Shopify, Klaviyo, Meta Ads, Google Ads, Recharge and 50+ other tools into one data warehouse that every agent reads from.

## Who Finsi Is For

Founders, CEOs and CFOs of e-commerce brands, from small brands to $200-400M in revenue. They know competitors are moving faster with AI, and their team is stuck doing manual, spreadsheet-heavy work before every decision.

Finsi is not about replacing people. The point is to let the team you have do more: agents handle the reconciling, reporting and first drafts, and people make the calls.

## The Problems Finsi Solves

We have talked to more than 200 e-commerce teams about AI. These problems come up in almost every conversation.

### 1. Decisions wait on spreadsheets
Someone exports Shopify, ads and Klaviyo, lines them up and reports once a month. Agents on Finsi do that work every day, so the team spends its time deciding what to do next.

### 2. Competitors move faster
Brands already running agents launch and adjust campaigns in days, not after weeks of meetings. The bottleneck one brand has, its competitor no longer does.

### 3. AI that makes up numbers
A general-purpose chatbot will invent a revenue figure. Agents on Finsi read from one set of definitions (a semantic layer), so revenue, margin and new customers mean the same thing everywhere.

### 4. AI copy that puts the brand at risk
A misspelled brand name in a campaign, or a health claim a supplement or CBD brand cannot make. The brand's rules and a human approval sit between every draft and customers.

### 5. Access nobody remembers granting
An employee's AI agent created Shopify API keys, and they still worked two months after that person left. On Finsi, access is granted per agent, recorded, and revoked in one place.

### 6. Agent know-how that leaves with one person
When the best agent lives on one laptop, it leaves when that person does. On Finsi, agents, prompts and their history belong to the company.

### 7. Teams whose agents never talk
Marketing's agent plans a discount on stock finance knows is running out. On Finsi, agents share context, so a campaign does not oversell inventory or blow the margin.

### 8. Lock-in to one AI vendor
Finsi works with models from several providers and picks the one that fits each task. If one provider raises prices or falls behind, agents can move to another, and the brand's rules, data and records stay where they are.

## The Three Rules Every Agent Follows

### Asks before it acts
An agent looks at the data and puts a specific change in front of the team, with its reasoning. Nothing changes money or customers until someone on the team approves. Finsi does not change ad spend unattended.

### Leaves a record
Every approved action is logged: what ran, who approved it and when, for every agent in one place.

### Stays inside its access
The brand decides, per agent and per brand, what an agent may read, what it may propose, and what it may run after approval.

## Data Foundation: One Set of Numbers

- **One data warehouse.** Finsi brings Shopify, Klaviyo, Meta, Google and subscription data into one warehouse, so agents are not each reading a different tool. Each company's data is stored in isolated, encrypted database schemas.
- **A semantic layer.** It defines what each number means, so "revenue" or "new customer" is the same thing for every agent and every report.
- **Metrics calculation.** Metrics are calculated once, with the same definitions across every source, and agents read from those instead of recalculating.

## The Platform

- **Connect your own agents.** Bring Claude, ChatGPT, Codex, Grok and others over MCP or API. They follow the same rules as the agents Finsi ships. Finsi MCP details: https://www.finsi.ai/ai/mcp
- **Data access controls.** Set, per agent and per brand, which data an agent can read and which actions it can propose.
- **Execution through Finsi.** An approved action runs through Finsi, not around it, so the approval and the record are always there.
- **Agents working together.** Put several agents in a group chat on one question and get one combined proposal back, not five separate ones.
- **Several AI providers.** Finsi uses models from several providers per task, so the brand is not locked to one AI vendor.

## Agent Library

Nine Finsi agents are ready to use, covering email, ads, creative, audiences, profit, search, customer research, competitors and reporting. The Agent Library puts them on one page: turn each on or off. They run under the same rules as any agent a brand brings.

- **ECHO (Email):** builds campaigns and flows in the brand voice and pushes them to Klaviyo.
- **PULSE (Ads):** proposes Meta and Google budgets and campaigns.
- **MUSE (Creative):** develops ad and email creative.
- **MAGNET (Audiences):** builds the segments and audiences campaigns target.
- **SCOUT (Profit intelligence):** reads the numbers to show where profit comes from and where it leaks.
- **BEACON (Search visibility):** watches how the store shows up in search and recommends fixes.
- **SAGE (Customer research):** turns customer feedback and survey answers into findings the team can act on.
- **RIVAL (Competitor watch):** watches competitors' prices, promotions, ads and emails and reports what changed.
- **Lead Analyst (Reports):** writes the weekly digest and reports, an optional daily brief, and answers questions about the numbers.

Running the work of an AI CMO (setting priorities, proposing budgets, reporting to founders) is one of the things agents on Finsi can do. It is not the whole of what Finsi is.

## What Agents Can Work On

These are the areas of the business agents on Finsi read and propose work in. Every proposed change still goes through approval.

- **Profit:** P&L, unit economics and contribution margin by channel, product, SKU and cohort.
- **Retention:** cohort analysis, churn signals and lifecycle insights for e-commerce and subscription businesses, with churn calculated on eligible cohorts.
- **Customer lifetime value:** LTV by acquisition channel, campaign, product and segment, with predictive LTV modeling.
- **Segmentation:** segments described in plain English, pushed to Klaviyo, Meta Ads or other platforms after approval.
- **Ads:** proposed Meta and Google budgets, campaigns, creative rotation and fatigue flags. Finsi does not change ad spend unattended.
- **Email:** campaign calendars, campaigns and flows written in the brand voice, pushed to Klaviyo after approval.
- **Attribution:** connecting acquisition spend to retention outcomes, not only first-order conversions.
- **Creative:** ad copy, headlines and creative concepts based on what has worked in the brand's own data.
- **Competitor monitoring:** tracking competitor ads and messaging shifts.
- **Support:** support ticket patterns and their link to churn.
- **Customer research:** survey design and analysis of open-ended answers.
- **Daily briefs:** what changed, what matters now and what to look at next.

## Integrations

Finsi connects with 50+ platforms via secure OAuth connections. Data syncs continuously, and historical data is imported on first connection.

### E-commerce Platforms
Shopify, WooCommerce, BigCommerce, and custom platforms. Order data, product catalogs, customer records, inventory levels, and storefront analytics.

### Subscription & Billing
Recharge, Skio, Stay.ai, Appstle, Bold Subscriptions, Loop, Smartrr, Stripe Billing, Chargebee, Recurly, Paddle, RevenueCat, Zuora, and FastSpring. MRR, ARR, subscription lifecycle events, dunning outcomes, and renewal patterns.

### Email & SMS Marketing
Klaviyo, Omnisend, Mailchimp, SendGrid, Customer.io, and Braze. Campaign performance, flow metrics, list health, deliverability data, and revenue attribution.

### Advertising Platforms
Meta Ads (Facebook & Instagram), Google Ads, and TikTok Ads. Campaign spend, creative performance, audience data, and conversion metrics.

### Analytics & Product Data
Google Analytics, Segment, Mixpanel, PostHog, Amplitude, and Heap. Behavioral data, conversion funnels, and product usage patterns connected to revenue outcomes.

### CRM & Customer Success
HubSpot, Salesforce, Pipedrive, ActiveCampaign, Close, and Copper. Customer records, deal pipelines, lifecycle stages, and communication history.

### Customer Support
Gorgias, Zendesk, Intercom, Freshdesk, and Help Scout. Ticket data, response times, satisfaction scores, and resolution patterns.

## How It Works

### Step 1: Connect your data
Link the e-commerce stack via OAuth: Shopify, Recharge, Klaviyo, Meta Ads, Google Ads, Zendesk and 50+ other platforms. Finsi brings it into one warehouse and calculates the metrics once.

### Step 2: Choose your agents and set their access
Turn on agents from the Agent Library, connect your own over MCP or API, or both. Decide per agent and per brand what each one may read, propose and run after approval.

### Step 3: Agents propose, your team decides
Agents do the reconciling, reporting and first drafts, and put specific changes in front of the team with their reasoning. A person approves each change before anything is written to Klaviyo, Meta or Google.

### Step 4: Approved actions run through Finsi
Approved actions execute through Finsi, and every one is recorded: what ran, who approved it and when.

## Humans When You Want Them

Forward-deployed operators from the e-commerce industry help set up and tune agents. Finsi Growth Agency is Finsi's growth team: it plans and executes on top of Finsi OS for brands that want it run for them.

## Results From Brands Using Finsi

Customers include Harney & Sons, 1906, Caffeine & Legends, History by Mail, PupBox, Vinyl Moon, Carolina Dream and Luda's Dumpling.

- **Caffeine & Legends:** MRR more than doubled in 6 months. https://www.finsi.ai/case-studies/caffeine-and-legends
- **Vinyl Moon:** +150 net new subscribers/mo, CAC down ~45%. https://www.finsi.ai/case-studies/vinyl-moon
- **Luda's Dumpling:** ~5x MRR over 11 months. https://www.finsi.ai/case-studies/ludas-dumpling

## How Finsi Differs From Analytics Dashboards

Tools such as Triple Whale, Lifetimely, RetentionX, Northbeam, Polar Analytics and Daasity focus on dashboards, attribution and reporting. A dashboard charts what happened. Finsi gives agents one set of verified numbers to work from and governs what they do next: they propose the step, a person approves it, and the action is recorded. Detailed comparisons:

- Finsi vs Triple Whale: https://www.finsi.ai/compare/triple-whale
- Finsi vs Lifetimely: https://www.finsi.ai/compare/lifetimely
- Finsi vs RetentionX: https://www.finsi.ai/compare/retentionx
- Finsi vs Northbeam: https://www.finsi.ai/compare/northbeam
- Finsi vs Polar Analytics: https://www.finsi.ai/compare/polar-analytics
- Finsi vs Daasity: https://www.finsi.ai/compare/daasity

## Company Information

Finsi, Inc was founded in 2024 and is headquartered in Jersey City, New Jersey. The company was co-founded by Andrei Rebrov and Misha Druzhinin. Finsi builds the place where e-commerce brands run AI agents under their own rules, so their teams can do more with the people they have.

## Contact

- Email: info@finsi.ai
- Website: https://www.finsi.ai
- Talk to us: https://www.finsi.ai/demo
- Blog: https://www.finsi.ai/blog
- Integrations: https://www.finsi.ai/integrations
- LinkedIn: https://www.linkedin.com/company/finlensai
- X (Twitter): https://x.com/FinlensAI