Data and applied AI for growing companies

Stop reconciling reports. One number for the whole business.

When your CRM, invoicing and operations give three different answers, every report has to be reconciled by hand. We define each metric once and, on that basis, AI answers business questions without making anything up.

Ask for a diagnostic See a real case →

info@landthor.com · we reply within 24-48 working hours.

Built on
PostgreSQLPostgreSQL
SupabaseSupabase
n8nn8n
MongoDBMongoDB
MetabaseMetabase
CloudflareCloudflare
GeminiGemini
ClaudeClaude

From source to decision

Each layer increases how much you can trust the data
01
02
03
04
05
Sources
CRM, ERP, operations, payments and marketing, each with its own numbers.
Bronze
Raw data, exactly as it arrives from each API, with its full history.
Silver
Normalised: a single customer identity and business rules written down.
Gold
KPIs ready for sales, finance and operations, with one shared definition.
Decision
Dashboards and AI on data management can actually rely on.
01 — The problem

When your CRM, invoicing and operations report different figures, every report has to be reconciled by hand.

We unify the sources and define each metric once, so every department works from the same figure.

Professional reviewing a data dashboard on a laptop
One set of figures everyone can rely on
02 — What we do

Three stages: connect, model, activate.

Separate tools → one data layer
CRM Invoicing / ERP Operations Payments Marketing Data layer normalised · versioned Management KPIs · IA
01

Connect

We connect your CRM, ERP, payments and operations through their APIs. Loading is automatic and daily: no more manual exports.

02

Model

We organise the data in three layers: raw data, normalised data with a single customer identity, and query-ready KPIs.

03

Activate

Management dashboards, automated alerts, encrypted backups, and documentation so your team can run it.

03 — DataQuick case

Three separate systems, one database.

We unified CRM, invoicing and operations through a Bronze, Silver and Gold architecture with shared business rules.

Read the full case →
Challenge
The CRM, accounting and operations reported different figures.
Solution
A central database with a single customer identity, the full history, and each metric defined once.
Result
The same figures across every department, with 225 automated tests checking them daily.

04 — Why Landthor

What you keep, beyond the dashboards.

A data project is judged the day it ends: by what stays running, by how much of it you understand, and by how free you are to carry on without whoever built it.

  1. You talk to the people building it

    There are two of us and we do the work ourselves. Whoever designs the architecture is the one who builds it and explains it to you. No junior profiles learning on your project, no layers of account management.

  2. The platform is yours from day one

    We work on your infrastructure wherever possible: your database, your repository, your credentials. If you decide to carry on without us tomorrow, you can.

  3. You get the manual, not just the result

    We document the decisions, the business rules and how each piece is operated. Handover is part of the project, not a separate line on the invoice.

  4. The data watches itself

    More than 200 automated checks verify every day that the numbers add up. If something arrives wrong from a source, we see it before you do.

  5. Nothing proprietary tying you down

    PostgreSQL, Supabase, n8n and Metabase. Open, standard technology, so any team can pick up the work without depending on us.

  6. Security and backups, by design

    Per-person permissions and encrypted backups outside the main platform. Defined before the first query is written, not once there is a problem.

05 — How we work

Three ways to start

Option 01

Data diagnostic

A short audit of your sources, data quality, business rules and automation opportunities.

Resultado
A map of sources, risks, quick wins and a roadmap.
Option 02

DataQuick implementation

The base data platform: ingestion, a three-layer model, KPIs, dashboards and documentation.

Resultado
An analytics layer built on your real data.
Option 03

Monthly support

Evolving KPIs, new sources, operational support, automation and AI on your internal data.

Resultado
Continuous improvement with no in-house data team.
See services, method and deliverables →
06 — Team

Two partners, and we do the work ourselves.

Whoever designs the architecture is the one who builds and documents it. No juniors learning on your project, no layers of management in between.

Documented architectureYour own repositoryHandover included
See how we work →

07 — Contact

Let's start with a diagnostic.

We look at which tools you use, where each piece of data lives and what can be automated, before proposing anything.

  1. 01
    You write

    Tell us which tools you use and what does not add up. A paragraph is enough.

  2. 02
    We reply within 24-48 working hours

    One of the two partners answers, not a salesperson.

  3. 03
    Half an hour on a call

    You leave with a map of your sources and the first three steps. No commitment.

Write to us at

info@landthor.com Ask for a diagnostic

We reply within 24-48 working hours.

There is no form and no public calendar: the email goes straight to the two partners and we arrange the call by email.