Italiano English Français Español Русский
Let's talk

AI automations · Examples

Examples of AI automations

Repetitive work we have taken off the hands of companies in different sectors. For each one: how it was before, what the automation does and how much time it gives back.

Tell us about your workHow we work

In short: here you will find 14 artificial intelligence automations that Dabryx has delivered for companies in trade, retail and fashion, construction and hospitality, as well as the ones we use every day in our own work.

For confidentiality we do not name our clients, only their sector. Every automation adapts to the programs the company already uses and can run in the cloud or locally, when data must not leave the company.

The gains shown are estimates, based on the assumptions written next to each example.

Sales and retailers

Delivered Sector: Fashion retail

Online retailers' prices and product pages under control

The problem
A brand sold by dozens of online shops and marketplaces doesn't know who is selling below the recommended price, who has wrong product pages or old photos, who has run out of stock. Checking by hand means opening hundreds of pages, and usually nobody does it.
What the automation does
Every day the automation visits the shops and marketplaces, compares prices, product pages and availability with the price list and sends a summary of the anomalies only: who is selling below price, which pages need fixing, where stock is missing.
What you gain
About 230 hours and €6,900 a year. Below-price sales are spotted the same day, not when a retailer complains.Estimate: 1 person, 5 hours a week, at €30 an hour of company cost.
Delivered Sector: Trade

Offers and quotes ready from an email or a voice message

The problem
Every request for an offer means looking up items, the price list and the customer's terms, then laying out the document. With many requests a week, offers go out the next day or even later.
What the automation does
The sales rep forwards the request or records a voice message ("offer for this customer, these products, these quantities"). The automation prepares the offer with the right items, prices and terms, and leaves it ready to check and send.
What you gain
About 310 hours and €9,300 a year. The customer gets the offer the same day, while interest is still high.Estimate: 10 offers a week, 40 minutes saved on each, at €30 an hour of company cost.
Delivered Sector: Trade

Orders from emails and PDFs loaded into the management system

The problem
Many customers still send orders by email, as PDFs or Excel files, each in their own format. Someone copies them by hand into the management system, line by line, with the risk of errors in codes and quantities.
What the automation does
The automation reads the order, recognises the customer, items and quantities, checks them against the catalogue and loads them into the management system. Doubtful lines, such as a code that doesn't exist or an unusual quantity, are flagged to a person before confirmation.
What you gain
About 245 hours and €7,400 a year. Fewer typing errors and orders shipped sooner.Estimate: 40 orders a week, 8 minutes saved on each, at €30 an hour of company cost.
Delivered Sector: Fashion retail

Lost sales in store reported to the right department

The problem
When a sale falls through in store, because a size is missing, a colour isn't liked or customers don't want a material, the information stays with the sales staff and is lost.
What the automation does
Store staff send a voice message. The automation understands the reason and forwards it to the right department: product issues to development, missing stock to the warehouse, which can already prepare the restock.
What you gain
About 500 hours and €15,000 a year. Data on missed sales that did not exist before, useful for purchasing and collections.Estimate: 10 stores, 10 minutes a day each to collect and pass on reports, 300 opening days, at €30 an hour of company cost.

Purchasing

Delivered Sector: Construction

Supplier quotes compared automatically

The problem
For every purchase, quotes arrive from several suppliers, each in its own format: PDF, Excel, text in the email. Lining them up to compare prices, delivery times and terms takes hours.
What the automation does
The automation reads the quotes, extracts unit prices, quantities, lead times and payment terms and puts them in a single table, highlighting the best offer and the items that don't add up.
What you gain
About 360 hours and €10,800 a year. Decisions are made with all the numbers in front of you, without missing the item hidden at the bottom of the PDF.Estimate: 20 comparisons a month, an hour and a half saved on each, at €30 an hour of company cost.
Delivered Sector: Trade and e-commerce

Marketplace prices, ratings and sales to decide what to buy

The problem
Deciding what and how much to buy requires data that sits outside the company: how a product's price changes, how its rating evolves, how much it has sold in recent weeks. Collecting it by hand from marketplaces is slow and rarely done.
What the automation does
Every week the automation downloads from the marketplaces the data on the products of interest and prepares a table with price, change over time, rating and recent sales, together with a short comment on what is worth buying.
What you gain
About 185 hours and €5,600 a year. Purchasing based on data updated every week, not on figures from six months ago.Estimate: 1 person, 4 hours a week, at €30 an hour of company cost.

Product and catalogue

Delivered Sector: Fashion retail

Customer reviews turned into reports for quality and product development

The problem
Product reviews are scattered across marketplaces and online shops, in many languages. They contain recurring defects and customer requests, but nobody has time to read them all.
What the automation does
The automation collects the reviews, translates them, groups them by product and by topic (fit, materials, defects) and sends quality and product development a summary of what keeps coming up.
What you gain
About 140 hours and €4,200 a year. A recurring defect shows up after a few reviews, not at the end of the season.Estimate: 1 person, 3 hours a week, at €30 an hour of company cost.
Delivered Sector: Trade

Data sheets and photos turned into catalogue data

The problem
For every new item, someone has to copy sizes, materials and features from the supplier's data sheet, then rename and sort the photos. With hundreds of items a year, that is weeks of work.
What the automation does
The automation reads PDF data sheets, extracts the data and prepares it for the catalogue or the PIM. It recognises the photos, links them to the right item and classifies them by type and colour.
What you gain
About 225 hours and €6,800 a year. New products reach the catalogue and the online shop sooner.Estimate: 300 new items a year, 45 minutes saved on each, at €30 an hour of company cost.
Delivered Sector: Hospitality

Descriptions, product sheets and messages translated into every language

The problem
Every text has to be translated into the languages of each market: descriptions, product sheets, manuals, messages to customers. With an external translator it takes days or weeks, and every change starts over.
What the automation does
The automation translates texts as soon as they are ready, keeping the company's terminology. For the most important texts, a native speaker only does the final review.
What you gain
About 265 hours and €8,000 a year. Texts ready in every language the same day, instead of weeks later.Estimate: 200 texts a year in 4 languages, 20 minutes saved on each translation, at €30 an hour of company cost.

Operations and management systems

Delivered Sector: Retail

One report every morning from systems that don't talk to each other

The problem
Sales, warehouse, purchasing and production live in different programs, developed at different times. To get an overview, someone extracts the data from each one and combines it in Excel, every day.
What the automation does
Overnight the automation reads the data from the different systems, combines it and at 8 am sends each manager their summary: what was sold, what is running out, what is late.
What you gain
About 230 hours and €6,900 a year. Everyone starts the day from the same numbers.Estimate: 1 person, 1 hour a day, 230 working days, at €30 an hour of company cost.
Delivered Sector: Trade

An assistant that answers questions about internal procedures and documents

The problem
Procedures, data sheets, certifications and manuals are kept in different folders. Finding an answer means searching for a long time or asking the colleague who knows, interrupting them.
What the automation does
Staff ask a question in their own language, as they would ask a colleague. The assistant searches the company's documents and answers, citing the document the information comes from.
What you gain
About 460 hours and €13,800 a year. New staff become independent sooner, experienced staff are interrupted less.Estimate: 10 people, 1 hour of searching saved each per week, at €30 an hour of company cost.
Delivered Sector: Construction

Voice messages from the field turned into warehouse requests

The problem
People on a building site don't open a management system to say materials are missing. Requests go through phone calls and messages, and they often get lost or arrive late.
What the automation does
The worker sends a WhatsApp voice message ("we're out of wall plugs and two rolls of membrane"). The automation turns it into a tidy request for the warehouse, which prepares the material and sends it to the right site.
What you gain
About 290 hours and €8,700 a year. Fewer wasted trips and fewer hours spent waiting for materials.Estimate: 5 crews, 15 minutes a day each on phone calls and follow-ups, 230 working days, at €30 an hour of company cost.

Communication and organisation

Delivered Sector: Software house

Emails routed to the right department, with a draft reply and reminders

The problem
Shared inboxes receive all kinds of requests: accounts, sales, support. Someone has to read them all and pass them on, and the unanswered ones risk falling behind.
What the automation does
The automation reads each email, understands what it is about and passes it to the right person, with a draft reply ready. If nobody has replied after a set time, it sends a reminder.
What you gain
About 305 hours and €9,200 a year. No customer request is left unanswered.Estimate: 4 people, 20 minutes a day each, 230 working days, at €30 an hour of company cost.
Delivered Sector: Software house

From meeting to summary, with everyone's tasks

The problem
After every meeting someone has to write the minutes and tell each person what they need to do. Often it doesn't happen, and decisions get lost.
What the automation does
The meeting is transcribed. At the end, the automation prepares the summary, decisions and tasks, and an email for each person involved, ready to send.
What you gain
About 90 hours and €2,700 a year. Everyone knows what to do the same day, and decisions stay in writing.Estimate: 4 meetings a week, 30 minutes saved on each, at €30 an hour of company cost.

How we calculate the gains

Each estimate starts from a simple assumption, written under the example: how many people do that task, how often and how long it takes. The hours saved in a year are multiplied by €30 an hour, an average company cost for an office employee.

They are orders of magnitude, not promises: in your company the numbers may be higher or lower. That's why, before proposing an automation, we measure with you how long the task takes today.

The estimates do not include the benefits that are harder to measure, such as errors avoided, faster replies to customers and information that used to be lost.

Learn more

Frequently asked questions

Can these automations be adapted to my company?

Yes. The examples show what we have already delivered, but every automation is built around the way the individual company works: its programs, its documents, its rules. We start from one specific task and extend it once it works.

Do they also work with old or custom-built management systems?

In most cases, yes. We connect to the programs in use when they allow it; otherwise we work with exports, shared files or emails, without having to change the management system.

Does the data stay in the company?

If needed, yes: the automations can run on an AI installed locally, on the company's own machines, so documents and data never leave. For non-confidential data, cloud models can also be used.

Where do we start?

With a conversation about the tasks that waste the most time today. Together we choose the first one to automate, measure how long it takes now and, once the automation is running, how much time it saves.

Which task would you like to get rid of?

Tell us in a few lines how it works today: we'll tell you whether and how it can be automated, and how much time it could give back.

Email

info@dabryx.com

Response times

Within 2 hours on business days (9:00–18:00) for clients with a support contract.

Custom software

Management software and custom software from €69 per month, with direct support from the developers 7 days a week.

AI consulting

Basic installation from €2,900, consultancy from €100 per hour, on your company's server or Mac.

Where we work

Throughout Italy: we come on site to the client's premises and also support you remotely.