Analysis templates are a ready-made library for working through your data: sales forecasts, customer segments, assortment, advertising, Wildberries and Ozon reports, and an analytics course. The library holds 257 templates: two groups for working with your own data — Business (9 categories) and Finance (10 categories) — plus a training course of 50 lessons. Instead of a blank page you open a ready-made analysis, press “Use this template” — and the agent adapts it to your data, runs the calculations and installs the finished report in Analytics Agent. You never see a formula or a single line of SQL: every template ends in a conclusion in plain language and a concrete action.
How it works
- the Analytics Agent sidebar has a Templates section — a live tree of the library: three groups, the categories, and the templates inside them;
- open a template — it carries a description in plain language: what you get, why you need it, how to run it, and how to adapt it to your data;
- press “Use this template” — the agent’s chat opens with a ready request: the path to the template and a request to apply it to your data;
- the agent reads the template, agrees with you on the data source, adapts the scripts, runs them and installs the finished report in the Documents section;
- the report is ready to open right away: it has interactive charts, tables and conclusions; the report header carries an “Ask agent” button for questions as you read.
Every template ships with demo data: you can study it end to end before it ever touches your own data.
Library categories
The templates are gathered into two groups by the task you are solving. Business is about the operational side of the trade: sales, customers, marketing and advertising, retail chains and selling on marketplaces. Finance is about money at any scale: from one family’s bank statements to a credit portfolio, a securities portfolio and a marketplace seller’s finances. Click a category card — you get the full list of its templates: what each one gives you and what it is worth.
Business
9 categories, 153 templates. What they are about: what to expect from next quarter’s sales, who your customers are and what they cost you, what drives a retail chain, where the ad money went and what it brought back — and two big marketplaces from the seller’s side, from the economics of a single sold unit to finding a niche.
Sales
Revenue forecast, customer segments, the market basket, customer value and churn — 10 templates for anyone who sells.
Marketplace Demand
Where orders leak away: the funnel, card conversion, returns, restocking, search and niche — 7 templates.
Customers
Customer value against acquisition cost, base analysis, loyalty and abandoned carts — 4 templates.
Marketing
End-to-end analytics, ROMI, attribution models, the site funnel and A/B tests — 8 templates.
Advertising
Campaign metrics, channel benchmarks, creative fatigue and incrementality checks — 8 templates.
Social Media
Channel metrics, the audience portrait, content, social ROMI and influencers — 8 templates.
Wildberries
53 templates built around WB specifics: unit economics, cards, buyouts, reviews, rates, taxes and competitors.
Ozon
48 templates built around Ozon specifics: unit economics, cards, buyouts, reviews, rates, taxes and competitors.
Finance
10 categories, 54 templates. What they are about: one family’s bank statements, the credit decisions and the loan book behind them, the share of fraud in the payment flow, the portfolio you hold and rebalance, the price of a bad day or a named crisis, a company’s health and worth from its reported numbers, forecast corridors for decisions, trading with its honesty rules, the macro numbers standing above every decision — and a marketplace seller’s finances.
Personal Finance
Where the money goes, the subscription audit, plan vs actual, the reserve and debts — 6 templates for personal money.
Credit Risk
Application scoring, default assessment, expected losses and portfolio quality — 5 templates for those who lend money.
Portfolio
The efficient frontier, risk optimization, fund comparison, rebalancing and retirement — 6 templates.
Market Risk
How much one bad day can take: checks on history, stress scenarios, volatility and the market’s “weather” — 4 templates.
Fundamental Analysis
A company’s health from its official reporting: ratios, the Z-score, valuation and insiders — 6 templates.
Forecasting
Revenue scenarios, cash gaps, subscription health, recession signals and news tone — 5 templates.
Trading
Honest strategy checks, a trade journal, a screener and correlations — 8 templates without gloss.
Macroeconomics
What actually moves markets: correlations, real rates and “green” returns — 3 templates.
Marketplace Finance
What actually stays from a sale: unit economics, break-even, buyouts, advertising and purchasing — 7 templates.
Training course
The third part of the library is a complete free course on working with data: six levels, 50 lessons. Every lesson is a template with demo data: ten minutes to complete, one skill that stays. The course runs from “what a table is” to a forecast with an honest corridor and a fair A/B test — and every comparison is made against the data’s own past and its own segments, never against “industry norms”. Click a level — you get all of its lessons and what you will learn in them.
Level 1. Data Basics
8 lessons: what tables are, where data comes from, sorting, filters, pivot tables and cleaning.
Level 2. Charts
9 lessons: bars and lines, the traps of pie charts, histograms, box plots, heatmaps and funnels.
Level 3. Data Processing
8 lessons: joining tables, dates and text, grouping, growth rates, the moving average and indexes.
Level 4. Dashboards
9 lessons: KPI cards, panels for operations, the manager and channels, design, filters and drill-down to details.
Level 5. Business Metrics
9 lessons: metrics and KPIs, revenue, conversion, retention, cohorts, customer value and ABC analysis.
Level 6. Advanced Analytics
7 lessons: correlation and cause, A/B tests, confidence intervals, the forecast corridor and your first project.
Searching the library
The Templates section has its own search: look by names, descriptions, tags and categories — for example, “forecast”, “buyout”, “NPS”. Results never mix with the documentation search.
Language and demo data
The library is bilingual: the Russian and English versions match exactly in content (257 templates in each). The language follows the installation’s brand — for Xedant Analytics that is English.
Demo data are real or plausibly generated datasets: from public SEC EDGAR and FRED data to synthetic worlds with planted patterns. That is why every template can be verified end to end: the forecast is checked against real numbers, and the honesty of the conclusion — against a truth known in advance.
What is under the hood
- a template is a folder: a description, Python scripts, demo data and finished reports;
- the scripts run not inside Analytics Agent but through the data agent (Xedant Data Agent) via its API; Analytics Agent itself executes nothing;
- calculation results land in the Lakehouse, reports in Analytics Agent documents, and alert rules (for monitoring templates) in the Alerts section;
- the library never changes in use: the template files stay untouched; the adapted scripts the agent creates are separate;
- the reports support interactive tricks: a whole-page filter by clicking a row or category, “[variable]” child pages and the “Ask agent” button in the header.
Monitoring templates (anomalies, metric control, channel health) ship with ready alert rules: the agent installs them as it applies the template, and the system itself reports deviations to Telegram, Slack or email.