Tenants of their own data
When agents instead of people work with the software, vendors change the rules for how they charge. Increasingly, customers pay per transaction and for access to their own data.
NXTGN AI · English translation
Read7 MIN
The Hoodie Club · by NXTGN
Welcome to the Hoodie Club
Our news and knowledge space where casual wear meets innovative thinking. It is about what matters to us: technology, engineering and the possibilities of digital transformation.
The hoodie has long stood for developer culture: casual and uncomplicated, yet full of complex ideas waiting to be explored. In that spirit, we share a look behind the scenes at NXTGN and knowledge from our work.
Knowledge & perspective
Which topics are relevant to you?
Search articles, briefings, whitepapers and technical deep dives. Filter by AI, Microsoft business applications, IoT or product management, from first orientation to project preparation.
28results
Hiding results is too late
A search across all repositories has to determine who is allowed to see what. The easy answer is to search first and filter afterwards. That approach has three gaps, none of which is obvious from the result.
NXTGN AI · English translation
Read5 MIN
Dataverse storage validation from 9 November 2026: stages, deadlines and options
From 9 November 2026, Microsoft will progressively restrict tenants that exceed their Dataverse capacity. Deadlines, affected tenants, options and a checklist.
Capacity Control · English translation
Read7 MIN
Microsoft storage, integration and AI: comparing SaaS and PaaS prices
Dataverse or Blob Storage, Power Automate or Logic Apps, Copilot or Foundry: list prices, the multiplexing rule and when each option costs less.
Capacity Control · English translation
Read12 MIN
Dataverse audit logs: growth, retention and safe cleanup
Why Dataverse auditing fills log capacity, what auditretentionperiodv2 does, and how to reduce audit storage without losing evidence.
Capacity Control · English translation
Read6 MIN
Keep sandboxes out of administration mode: 11 cleanups without buying more capacity
Cleanups that free Dataverse storage without additional purchases, with risks, effort and navigation paths. Only measures documented by Microsoft.
Capacity Control · English translation
Read9 MIN
A confusion of languages
In established businesses, the same thing has different names depending on the department, location and generation. What that costs in everyday work, and why a shared language is a leadership responsibility.
NXTGN AI · English translation
Read5 MIN
Everything digitised, nothing findable
All the files are there, the search is running, and still nobody finds the answer. Why full-text search in a business reaches a limit that is not obvious from the results.
NXTGN AI · English translation
Read4 MIN
When you need a large AI model — and when you do not
The largest general-purpose cloud models are often seen as the quickest route to AI in a business. For questions about the company’s own operations, they do not solve the harder part.
NXTGN AI · English translation
Read6 MIN
Why we do not translate
The manufacturer’s documentation is in English, but questions are asked in German. Translating it would be the obvious approach. At NXTGN, we decided against it.
NXTGN AI · English translation
Read5 MIN
Artificial intelligence creates artificially intelligent people
Why fluently worded AI answers can bypass our judgement, and the difference between using a tool and knowing how to operate it.
NXTGN AI · English translation
Read5 MIN
Which AI use cases are actually worth pursuing?
Not every task involving text or data needs AI. A useful case combines a real problem, suitable information, measurable value and manageable error risk.
Read9 MIN
Why CRM projects rarely fail because of the software.
Dynamics 365 can support many sales, service and marketing processes. Everyday success depends on process clarity, data, ownership and adoption.
Read10 MIN
IoT business case: seven questions before the first sensor.
Connectivity and hardware matter, but they are not the starting point. A viable connected product begins with customers, value, economics and the process improved by device data.
Read11 MIN
ZONIVO shows what a cleanroom reading means.
A differential pressure reading says little on its own. What matters is the room it was measured in, whether the value was valid and which batch was running at the time. We developed ZONIVO for these questions. It is our new brand for cleanrooms.
Read5 MIN
Copilot is not one product. It is a product family.
Microsoft uses the Copilot name for assistants, embedded features and agent platforms. This briefing provides a practical map and identifies the decisions companies actually need to make.
Fusion IT: when business and IT deliver together.
Low code, data platforms and AI change not only tools but responsibilities. Fusion teams combine domain expertise, product ownership and professional engineering.
Product manager: the role connecting digital transformation.
Digital solutions are not finished at go-live. Someone must continuously connect user needs, business value, technology, data and operations.
Company knowledge with AI: knowledge that can work in the process.
A company knowledge system is not another search box with chat. It is a knowledge architecture connecting sources, permissions, retrieval, models, evaluation and real work.
Dynamics, Fabric & Foundry: which platform does what?
Dynamics 365, Power Platform, Dataverse, Fabric, Power BI, Copilot Studio and Foundry form an architecture for operations, data, analytics and agents when responsibilities are explicit.
AI operating model: from solution to accountable operations.
Productive AI needs a delivery and operating model connecting use cases, models, knowledge, agents, approvals, evaluation, cost and incidents.
PPWR and documentation of IBC cycles
ibc.digital explains how to document IBC cycles using an app, QR codes and tracking. Original guide in German.
ibc.digitalOriginal article in German
Read (opens in a new tab)ON IBC.DIGITAL
Re:Think field service
From break-fix service to remote support, connected field service and predictive maintenance.
Read7 MIN
From sensor to business model
How market, revenue model and customer process shape a connected-product architecture.
Read8 MIN
Company knowledge with AI for mid-sized companies
RAG, GraphRAG, permissions, evaluation and model routing for trusted company knowledge.
Read9 MIN
Local AI, open weight and frontier
Decision criteria for data protection, quality, latency, cost and operations.
Read8 MIN
Connect CRM and ERP deliberately
Systems of record, shared master data and clear process boundaries without another data shadow.
Read7 MIN
Run digital solutions as products
Connect roadmap, users, business model, platform and operations beyond the first go-live.
Read7 MIN
Formats
Read, download and soon listen.
Choose a quick explanation or a detailed deep dive depending on your question. Podcasts and videos will join the Hoodie Club in a later release.
Articles & explainers
Terms, decisions and common project questions explained without product brochures or buzzword collections.
Briefings & whitepapers
Detailed material with decision paths, architecture, roles and checklists to read or download.
Podcasts & video
Expert conversations, short explainers and a look into technology, engineering and our project work.
Behind the scenes
Technology becomes interesting when it is understandable and useful.
From new developments in software and AI to architecture and practical experience with business applications and IoT, the Hoodie Club is a place to learn and think further. We write about what we build, test and encounter in projects, including where a technology is not ready yet.
Join us in the world of code, products and digital transformation and discover what makes the work at NXTGN different. Welcome to the club!
The Hoodie Club
Which topic are you missing?
Send us your question. If it is relevant to other clients too, we may turn it into one of our next articles or deep dives.