Fast market entry thanks to scalable data annotation
Benefit from precise data preparation by an international team with a social mission.Machine learning is already being used in many areas: identification of road users for autonomous driving, detection of diseases in health care, aerial image and earth observation data analysis, recognition of handwritten documents in business processes, and in cultural heritage archives. Humans are indispensable in the development of artificial intelligence. They must perform the necessary preliminary work by manually marking shapes, patterns, movements and objects in image data so that the software can then learn from this information. Manual preparatory work is very time-consuming and therefore expensive.
We can do this work for you at a reasonable price. Thanks to our impact sourcing partner Digital Divide Data, who runs operations in South-East-Asia and Africa based on a social mission, we are able to index your data quickly and reliably, with consistently high quality and at a fraction of European labour costs. We can flexibly adapt our capacities to your data volume, which will give you a lower entry barrier for feasibility studies and a faster time-to-market. Read the interview with founder Daniel Lanz.
- Mail us: info@lanz.ai
- Telephone service: +49-2207-9610-550
The solution your business needs.
Consulting
- We help you select, create or purchase suitable data and jointly define the necessary metadata standards as well as pre-processing and normalisation steps.
- We advise you on data annotation methods, training metrics and evaluation strategies; thanks to our experience in model training, we support you where data seamlessly transitions into machine learning.
- Areas of application include computer vision and data analysis for applications such as production control, agriculture monitoring and transport management.
Data Annotation
- We perform manual object recognition and segmentation on image and video data as well as tagging of audio and text to create structured annotation data.
- Our workflow includes specification, piloting, scaled annotation and quality assurance - a well established process.
Tagging
- Our tagging service offers customised tagging - from images, objects and metadata to text and audio data - using common standards such as Dublin Core, MODS, METS/ALTO or ISO 11179.
- We design structured taxonomies and guidelines to ensure consistency and findability.
Geospatial Data Services
- Together with our partner Regio, we deliver end-to-end geospatial data services - from data capture to system deployment.
- We collect spatial data from satellite, drone, aerial, and sensor sources, and process it into structured GIS datasets through digitizing, georeferencing, classification, and integration with tabular information.
- We design spatial databases and infrastructures, develop GIS systems and data workflows, and provide standards-based APIs (WMS, WFS, WCS) for seamless platform integration.
- We deliver tailored spatial analysis and interactive visualizations—such as maps and dashboards - to support decision-making in areas like infrastructure, environment, transport, and public sector planning.
- We offer full infrastructure deployment and support, using open-source platforms such as PostGIS, QGIS, and GeoServer, as well as commercial alternatives, depending on project requirements.
AI-based Data Analytics
- By pre-processing (e.g. noise removal, format conversion), tagging and analysing complex data sources, we ensure a holistic view of your data treasures.
- Our tools and processes - including automatic tagging approaches - are designed to efficiently extract patterns and insights from large volumes of data.
Computer Vision Models
- We train, evaluate and optimise machine learning models (e.g. CNN, YOLO, U-Net) for image and video data for applications in various industries, for example logistics, insurance or security.
- Our team supports you in the selection of suitable training methods, metrics and pipelines and accompanies you through the entire evaluation process - both in an advisory and operational capacity.
A
network you can trust.
About LANZ GmbH
Over 25 Years of Expertise in Digitalisation and AutomationThe company was founded in 2018 by Daniel Lanz, who has over 20 years of professional experience in digitization, data acquisition and software automation. Previously, Lanz was long-time CEO at Hamburg-based software and digitization company Content Conversion Specialists, Managing Director at communications agency Neonred and Director Business Development at data services provider Digital Divide Data.
- Mail us: info@lanz.ai
- Telephone service: +49-2207-9610-550
FAQ: What You Should Know.
We're here to help.
+49-2207-9610-550Data annotation (also known as "tagging" or "labelling") refers to marking objects or patterns in raw data (e.g., images, audio, or text) to provide machine learning models with clear training data and correct labels. Without a high-quality labelled dataset, AI accuracy suffers significantly.
Typical phases include:
a) Defining the specification – What labels are needed and what should they look like?
b) Pilot phase – Test annotation to review and fine-tune.
c) Scaled annotation – Main annotation following the final specification.
Especially in smaller projects, one third of the time is spent on clear guidelines and testing.
In-house annotation (e.g., by working students) is time-consuming and often leads to inconsistent quality. Specialized providers like LANZ.AI usually deliver better results thanks to trained annotators, optimised workflows, and proven methods.
LANZ.AI collaborates, among others, with Digital Divide Data – a social enterprise employing young people in Southeast Asia and Africa. This creates high-quality annotation solutions with social impact.
Not necessarily. Often, a clear and unambiguous specification is sufficient. Annotators do not need deep expert knowledge – in many cases, precisely formulating the annotation rules is the key.
AI is becoming increasingly important across all industries. Anyone working with large datasets will benefit, for example in:
• Quality control (e.g., detecting production defects)
• Agriculture (e.g., irrigation, pest infestation via satellite images)
• Maintenance/Inspection (e.g., pipe systems in chemical plants)
• Transport, healthcare, public administration, cultural heritage
• Expertise: Over 20 years of experience in specification, data governance, and workflows.
• Efficiency: Scalable, structured processes and fast execution.
• Social commitment: Impact sourcing for fair working conditions.
• Flexibility: Option to perform projects in Europe or locally in Germany for particularly sensitive data.
Yes – in addition to image data, LANZ.AI also annotates video material, audio data, text corpora, as well as metadata using structured label standards (e.g., Dublin Core, MODS, METS/ALTO).
• Annotation tools & attention to detail
• Metadata standards & management: Knowledge of established schemas such as Dublin Core, MODS, METS, ALTO, or ISO 11179
• Data preprocessing: Data cleaning, formatting, noise removal
• Data normalization: Consistent categories, formats, scaling
• Quality assurance: Peer reviews, inter-annotator agreement, audits
• Domain expertise or clear specifications: Guidelines created jointly with LANZ.AI form the basis for consistent annotation.
Yes – the LANZ.AI development team supports clients both consultatively (training, metrics, evaluation) and operationally (setting up training pipelines, quality assurance, model optimisation).
Let's build the future together.
