Skip to content
OrtmannConsulting

Automation · AI · Communications

Processes that run themselves.

I automate the repetitive work in sales, service and communications — using artificial intelligence where it measurably saves time. In production in weeks, not quarters.

14Documented projectsautomation, AI, communications
4 yearsLongest system lifespangrown in production
0Unevidenced figuresevery number comes from the project itself

01Services

Three fields, one goal: less manual work.

No innovation theatre. Every project moves a number you already measure.

01Automation

Workflows without the handoffs

Repetitive work in sales, service and administration — automated end to end, handing back to a person exactly where judgement is actually required.

  • Turn inbound email enquiries into draft quotes automatically
  • Check invoices and delivery notes without manual entry
  • Reports that build and send themselves
  • Keep master data in step across ERP, CRM and shop

02Artificial intelligence

AI that lands in daily operations

Language models applied where they demonstrably save time — as part of the working process, not as a demo for the board.

  • Classify and pre-answer documents and email
  • Knowledge bases people actually use
  • Check quotes and copy for completeness
  • Pre-qualify technical enquiries automatically

03Enterprise communications

Communication that does not die in an inbox

Internal and external communication made structured, traceable and automated — from a multi-country contact centre through to the channels your company already runs on.

  • Contact centre platforms across multiple countries and channels
  • Customer data from ERP and CRM in the agent workspace
  • Teams, Outlook and ticketing wired together properly
  • Escalation paths that are measurably honoured

02How it works

The way I work

Every stage is time-boxed. If the pilot saves nothing, we stop there — that is the point of splitting it up.

01

Analysis

1 week

We look at the process as it actually runs, not as it is documented. You end up with effort, saving potential and risk per step — and a straight recommendation about what is worth doing and what is not.

02

Pilot

2–3 weeks

One part of the process goes live, with real data and real users. Fixed price, so you know the downside before you commit.

03

Rollout

4–8 weeks

Extension to the full process, including training, documentation and handover to your IT team. After this you can run it yourselves.

04

Operation

Ongoing, optional

Monitoring, adaptation to new cases and regular review — or you take it over entirely. No lock-in, no forced maintenance contract.

03Work

Selected projects

Each one with the problem, the approach, and a number.

Group compliance, construction chemicals

2026
10,872Documents generated

10,872 EU declarations of conformity, not copy-paste

New EU packaging regulation required declarations of conformity in 12 languages for around 900 packaging materials. Instead of filling them in one by one, a pipeline generates and files them in a single run.

  • Python
  • docxtpl / Jinja2
  • Microsoft Word COM
  • Microsoft Graph API

Customer service, construction chemicals

2026
114 → 18Widgets in the call flow

From a 114-widget phone menu to a voice assistant

Callers say what they need in their own words instead of tapping through three menu levels. The language model is never allowed to name a destination — it picks from a list, and the phone system dials.

  • TypeScript / Node.js
  • Azure OpenAI
  • Twilio ConversationRelay
  • Azure App Service

04About

Alexander Ortmann

For more than twenty years I have worked where process meets system — and I have deliberately never picked a side. I can walk a quoting workflow through with the sales team and then build the integration for it myself. The systems have changed names several times in that period. The problems behind them have not.

What I have learned along the way: automation rarely fails on the technology. It fails because nobody knows the whole process. There is almost always a step somebody added years ago for a good reason — and nobody remembers the reason. Automate that step without understanding it and you have automated a mistake, and made it faster.

So I start with the process, not the tool. Before anything gets built, it has to be clear what the workflow actually does, where it stalls, and what happens when the automation gets it wrong. That is unglamorous and takes about a week — and it is the difference between a system that lands in daily use and one people quietly work around.

Experience
20+ years
Based in
Stuttgart
Languages
German, English
Systems
SAP, Microsoft 365, Azure, Salesforce, Twilio

05FAQ

What you are probably wondering

What does a project cost?

The pilot runs at a fixed price, so you know the downside before you commit. Everything after that is billed by effort or as a package — you get the estimate before the work starts, not after.

What happens to our data?

Processing inside the EU wherever possible, under a data processing agreement. Where language models are involved, we agree in writing beforehand which data they see and which they do not. If a process touches personal data, that gets settled before a line of code is written.

Does this work with our SAP or Microsoft 365?

Usually yes — both are the most common starting point in this market. During the analysis week we check which interfaces actually exist rather than promising it up front.

How long until we see something?

Two to three weeks to a working pilot. Not to a concept — to real cases running through it.

Will we be dependent on you afterwards?

No. Documentation, source code and handover to your IT team are part of the rollout. Ongoing support is an offer, not a condition.

What if we do not know where to start?

That is the normal case, and exactly what the analysis week is for. You come out of it knowing which three processes are worth doing, and in what order.

06Contact

Let's talk about your process.

30 minutes, free, no sales pitch. You describe what is costing you time — I tell you honestly whether automation is worth it.

Or email directly: Alexander.Ortmann@ortmann-consulting.de