You see whether the strategy works.
Specialty line sales are up 14% YoY, but its share of the portfolio is stuck at 18%.
You know this line’s messaging needs a closer look.
New · Smart assortment module
Your product range, groupedthe way you think about it.
This tool is for e-commerce people: online shop owners, managers and agencies. If you do not sell online, there is little here for you.
Smart assortment turns the descriptions you already have into product attributes: format, weight, roast level, origin. Where your shop sees whole_bean_1kg and “uncategorised”, you see shares, margins and gaps in your range.
89.5% of the catalogue after the first run
7 attributes
manual corrections apply instantly
full version on subscription, not in the trial
Catalogue coverage89.5%
Attributes✦ 7
Product format
5 values · 100% read
7 days · vs previous period
ValueShare of revenueRevenueMarginConv.
whole bean€22,10342.6%51%
capsules€4,626.5036.1%63%
ground€2,981.7540.8%31%
drip / sachets€962.7544.2%38%
tasting set€372.5047.5%97%
AI is describing the catalogue:548 of 612 products
89.5%
64 awaiting description
01 · First, tidy up the names
The “type” field today: 13 names for 5 product formats.
After the AI read: 5 values and 99.4% of revenue in the right place.
Keep scrolling: the AI is tidying up right now.
whole bean71.2%
capsules14.9%
ground9.6%
drip3.1%
set1.2%
products described548 of 61289.5%
02
From product description to figures, in four steps.
The module works on data DataOrganizer already pulls via API. It changes nothing in your shop, and the final say is always yours.
Step 1 · reading
AI reads product descriptions
It extracts values for seven attributes, from product format to origin. “Whole bean coffee Brazil Santos 1 kg, dark roast” becomes a complete set of attributes rather than a jumble of words.
Step 2 · honesty
Uncertain reads get a status, not an invented value
When a description cannot be read with confidence, the product goes to “to review” or “awaiting description” and is excluded from shares and conclusions.
AI does not invent valuesStep 3 · figures
Every value gets revenue, margin and conversion
Revenue, margin, units, conversion, share and trend for the active range from the last 6 months. Plus a matrix of two attributes at once.
Step 4 · your input
Correct with one click. Your correction takes priority.
Changing a value for a specific product applies instantly, with no catalogue recalculation and no credits. You can also swap an attribute for a different one.
corrections apply instantly03
What will you use it for?
Instead of digging through the whole shop, you look where something is happening.
Thousands of SKUs reduced to a handful of groups you already think in. You see which group stands out and start your analysis there. No need to go through all of the shop’s data.
Key benefit
You know where to dig deeper
One glance shows which group is growing, which is declining and which depends on discounts. You drill into the data worth exploring and spend no time on the rest.
You spot what depends on discounts.
Capsules: 19.3% of revenue is discount, against an average of 14.1%.
You know which price to review.
You notice a gap in the range.
There are no light roast capsules in the range, while light roast beans are growing fastest.
You know where to look for a new product.
You see the dead tail.
40% of SKUs (245 of 612) generate 4% of revenue.
You know which group to review product by product.
Any group that interests you expands into a product list. You can also export it for ads, Merchant Center or a spreadsheet.
Product listAdsMerchant CenterSpreadsheet
Other signals you will catch
They browse but do not buy.Drip: 18% of views, 3% of revenue.You know which product pages to review.
They drop off at the basket.Ground: basket → purchase 31% against an average of 50%.You know where to look for the obstacle.
High revenue, low profit.Dark roast: highest revenue, lowest margin.You know where to check cost and price.
Many SKUs, weak results.500 g: 30 SKUs, revenue per product 3× lower than 1 kg.You know where to look for redundant variants.
Units are not value.250 g: 51% of units, 32% of revenue.You know where to review price and basket.
The shop and marketplaces sell different things.Capsules: 12% in the shop, 31% on marketplaces.You know which channel offer to review.
A silent decline.Shop +9%, medium roast −18% YoY.You see it before it disappears in the overall total.
They buy several at once.Capsules: 2.8 units per order, whole bean 1.2.You know where to look for bundle ideas.
04
Your product data now speaks your language.
Attributes form a shared vocabulary: you see it in the tables, and the AI assistant in DataOrganizer and your chat via MCP understand it too. Ask about beans, roast or margin and you simply get an answer.
Which format really makes money?
Whole bean: 71.2% of revenue at a 42.6% margin. Capsules have the best conversion (63%) but sell mainly on promotion: discounts eat 19.3% of their revenue.product format · 7 days
What is missing from the range?
Light roast capsules. The capsules × light roast cell is empty, while light roast beans are growing fastest in the whole matrix.matrix · format × roast
Does 250 g cannibalise the kilogram?
No. The 250 g size attracts new customers, while kilogram buyers are mostly returning customers. The two values serve different groups.Customers tab · new and returning
These are not pre-recorded answers. The assistant calculates them from the same attributes you browse in the tables, so you are always talking about the same figures.
The bigger picture
Smart assortment is one feature.All of DataOrganizer works underneath.
Real ROAS from shop sales
9 ready-made dashboards
GA4 + shop purchase funnel
Product margin
✦ Smart assortment · you are here
Shop vs marketplaces
AI assistant in your language
MCP: Claude · ChatGPT · Gemini
Sheets and CSV exports
Compare any periods
Prices and discounts
Words from product names
19 integrations via official APIs · data refreshed daily · from €24 per month per source
05
It all starts with connecting your data.
DataOrganizer is a browser-based application that has been combining e-commerce shop data for many years: orders, marketplaces, GA4 and ad spend all go into a single data bank, calculated the same way. No implementation project and no changes to your shop’s code.
Explore the full tool
Shop (WooCommerce, Shopify…)
Marketplaces
Google Analytics 4
Google · Meta · TikTok Ads
9 dashboards
AI assistant and MCP
✦ Product attributes
Exports and reports
19 integrations via official APIs, no plugins: you connect them yourself in about 15 minutes.
The system fetches and combines data every day: the same periods, the same products, one version of the truth.
From €24 per month per source. Users, dashboards and the AI assistant included.
06
What else the same account can do.
Everything below is included in the account and works on the same combined data. No separate add-on modules to buy.
Real ROAS
Calculated on shop orders, not in the ad platform. No double attribution.
9 dashboards
Summary, products, customers, marketing, traffic. Ready from day one.
Purchase funnel
Sessions, basket, payment and purchase: GA4 and the shop in one flow, day by day.
Product margin
Purchase cost, shipping and returns for every product. Revenue stops pretending to be profit.
Shop vs marketplace
The same item in two channels, with fees included. You know where it pays to sell.
AI assistant and MCP
Ask in plain language in the app or in your own chat: Claude, ChatGPT, Gemini.
Exports
Google Sheets, CSV and data for your own analyses. Your numbers are never held hostage.
✦ Smart assortment
Product attributes from descriptions. This is the module you are reading about.
this page07
Who uses it every day.
Owner, marketer, agency and analyst: each looks from their own angle, but they all see the same numbers. No SQL, no training and no limit on accounts.
Shop owner
In the morning, over coffee, in five minutes: how much you earned and whether the ads pay off.
See what you gainE-commerce manager
A weekly report from a single view instead of an evening spent stitching exports together.
See what you gainPerformance agency
All clients on one standard. Real ROAS without asking for exports.
See what you gainAnalyst
A clean, combined database instead of building ETL. Sheets, CSV and MCP for your own models.
See what you gain08
Eight questions asked at every demo.
Collected from conversations with clients. Two-sentence answers, no accordions and no clicking.
What exactly is DataOrganizer?
A browser-based application that combines data from your shop, marketplaces, GA4 and ads into one data bank: ready-made dashboards, real ROAS, an AI assistant and exports. Smart assortment is one of its modules.
Where do I start?
You create an account and connect your sources via API, on your own, in about 15 minutes. For 14 days you test it without a card, and during that time AI describes 500 of your products, so you see the module on your own data. After that you pay from €24 per month per source.
Where does AI get the attributes from?
From the descriptions and product data that DataOrganizer already fetches from your shop via API. It changes nothing in your shop: the attributes live in DataOrganizer.
What if a description is too weak?
The product gets the status “to review” or “awaiting description” and is excluded from shares and insights. AI does not guess values.
Which products are described?
The active assortment from the last 6 months, because that is what builds the rankings.
Can I correct the AI reading?
Yes, with one click. Your correction overrides the AI reading and takes effect immediately, with no recalculation and no credits. You manage values and attributes in the settings.
Does it change anything in my shop?
No. The module only reads data. Your shop, descriptions and categories remain untouched.
How much does it cost and when does it work?
During the trial, AI describes 500 products free of charge. The subscription covers the whole active assortment: descriptions draw on a credit allowance granted with the subscription, which is enough for most shops. Browsing, rankings and manual corrections are always free.
You already have the descriptions. All that is missing are the attributes.
You create an account in a minute, connect your sources in a quarter of an hour, and DataOrganizer builds a data bank and dashboards from them. The assortment module runs on the same data: you get your first rankings the same day.
No credit card required
Cancel in 1 click
Support included
Smart assortment is available in full on accounts with an active subscription. The trial period is for getting to know the tool and connecting your data.