Forecasting by feel
Sales promises one number, purchasing orders another, and the budget says something else entirely. Nobody measures which forecast was actually right.
Forecasting · Inventory · S&OP · Automation
Your supply chain.
Finally predictable.
I help mid-sized manufacturers, distributors and retailers see demand coming, right-size their inventory and automate the reporting in between — with methods proven in the semiconductor industry, where planning mistakes cost millions.
- Fixed-price packages
- First results in 2–3 weeks
- You own everything I build
The starting point
Sound familiar?
Most mid-sized companies run excellent products on gut-feel planning. It works — until it doesn’t.
The wrong inventory
Cash sits in slow movers while bestsellers run out of stock. Safety stocks were set years ago — by someone who has since left the company.
Reporting that eats the week
Every Monday someone copies numbers between spreadsheets. By the time the report is ready, the decision has already been made — on opinion.
None of this needs a new ERP system, a data science team or an “AI transformation”. It needs the right methods, applied pragmatically. That is exactly what I do.
Services
What I do
Four building blocks. Take what you need — every engagement starts with a clearly defined, fixed-price scope.
Demand Forecasting
A measured baseline forecast for every product — improved month over month.
- SKU-level statistical forecasts, machine learning only where it pays off
- Forecast accuracy and bias tracking (MAPE, forecast value added)
- A monthly demand review your team actually runs
Purchasing decisions based on numbers, not nerves.
Inventory Optimization
The right stock in the right place — and slow movers off your balance sheet.
- ABC/XYZ segmentation with service-level targets
- Safety stocks and reorder points from real demand variability
- Excess and obsolete stock surfaced with € impact
Cash freed up without putting availability at risk.
S&OP & S&OE
One plan everyone works from — decided monthly, controlled weekly.
- A lightweight monthly S&OP cadence sized for mid-sized teams
- Weekly S&OE control: orders vs. plan, exceptions first
- Clear roles, clear decisions, one set of numbers
Fewer firefights, faster decisions.
Reporting, AI & Automation
Your KPIs refresh themselves — straight from your ERP and spreadsheets.
- Automated dashboards in your stack: Power BI, Excel or web
- Python and SQL pipelines that replace copy-paste rituals
- Pragmatic AI where it demonstrably pays back
Hours back every week — and decisions on live data.
Approach
How we work together
No 200-page strategy deck. We start small, prove value fast and only scale what works.
- 01
Intro call
30 min · freeYou bring your pain points, I bring questions. We find out whether and where I can help — and you take away first pointers either way.
- 02
Supply Chain Health Check
2–3 weeks · fixed price Start hereI dive into your data and processes and show you, in plain numbers, where money is sitting and what to do about it.
You receive:
- Forecast accuracy baseline by product family
- Top overstocks and understocks, quantified in €
- Safety stock and reorder point recommendations
- A prioritized 90-day roadmap
Your data stays yours — NDA as standard.
- 03
Build
4–8 weeksWe implement what the roadmap says: forecast models, inventory logic, automated dashboards — in your systems, documented, without license fees. You own every line of it.
- 04
Run & improve
optionalI stay on board as your planning co-pilot: monthly S&OP support, model tuning, new use cases — a few hours per month, for as long as it pays for itself.
Why me
Trained where supply chains are hardest
MZ Michael Zemler Supply chain analyst, semiconductor industry
By day, I work as a supply chain analyst at a global semiconductor company — an industry with lead times beyond 20 weeks, thousands of products and zero tolerance for guesswork. Forecasts, inventory targets and S&OP decisions there move millions, and the methods behind them are measured, challenged and improved every single month.
Here is what I keep seeing: those methods are exactly what most small and mid-sized companies are missing — not because they are secret, but because nobody has ever translated them to fit. That is what I do. No enterprise software, no army of consultants, no black boxes: pragmatic tools and processes, built in your systems, explained until your team runs them without me.
“If a method survives semiconductor planning, it will work for your business.”
Enterprise methods, SME fit
S&OP, S&OE, statistical forecasting and inventory management — practiced daily in one of the most demanding planning environments in the world.
I build, not just advise
Python, SQL, Tableau, n8n, Excel — you get working tools and automated reports, not a slide deck with recommendations.
You own everything
Every model, every dashboard, every line of code runs in your stack and stays with you. No license fees, no lock-in, no dependency on me.
Straight talk
If something will not pay back, I tell you before you spend money on it. The goal is your independence, not my utilization.
FAQ
Fair questions, straight answers
We’re not a data-driven company. Is this for us?
That is the typical starting point, not a problem. If your planning lives in Excel today, we start exactly there — the first improvements usually need nothing more than your order history and a structured look at it.
Which ERP systems do you work with?
I am system-agnostic: SAP, Microsoft Dynamics, Odoo, proprietary systems or plain Excel exports all work. What I need is your sales and order history — almost every system can export that.
Do we really need AI for this?
Only where it pays. A clean statistical baseline forecast and a working process often beat any sophisticated model. I will always show you the simplest thing that works — and the business case before any “upgrade”.
What does it cost?
Fixed prices, agreed before we start — beginning with the Health Check. No open-ended day rates, no surprise invoices. If I don’t believe a step will pay for itself, I will tell you.
How do you handle our data?
NDA as standard, GDPR-compliant processing, and wherever possible your data stays in your own systems. This website itself runs without cookies or tracking — that is how I like data to be treated.
Will this replace our planners?
No — it makes them better. The goal is a team that runs a sharper process with better tools, not a company that depends on an external consultant.
Let’s look at your numbers.
Thirty minutes are enough to find out whether — and where — there is money on the table for you. Free, no strings attached.