Payroll Processing: Excel + Python

A bridge between your shift data and payroll systems like DK or H3 — with algorithms tailored to your company, not the other way around.

Excel Interface Python Notebooks DK / H3 Ready Output Local or Cloud Tailored to your needs
What the algorithm takes over:
Manual validations
holiday pay · sick-leave · overtime
Transcription errors:
0
automatic validation
View Full Presentation

Detailed system workflow

No risk — you pay only when it works

We believe in the solution. That's why the payment model is simple: if we successfully automate your payroll processing, you pay. If not — you pay nothing.

Pay only on success

If payroll automation succeeds — you pay. If we don't achieve it — no charge.

Zero implementation risk

The Python code doesn't touch your existing payroll systems. It's purely a bridge — it doesn't interfere with DK, H3, or anything else.

You decide whether to use it

Accounting can choose to use the bridge each pay period or not. Your existing workflow is always there as a fallback.

How this works in practice

What the Python code does:

  • Reads shift data from your existing source
  • Performs calculations that accounting did manually
  • Returns Excel ready to import into DK/H3

What the Python code does not do:

  • Does not write to DK, H3, or any database automatically
  • Does not modify anything in existing systems
  • Does not require abandoning existing workflows

Think of it as an extra tool on the accountant's desk — not a system that replaces what's already there. If it saves time and works correctly, use it. If not, the old way is always still there.

Algorithms tailored to your company — not the other way around

With most off-the-shelf payroll systems, the company must adapt to the software: change workflows, conform to standardized rules, and forget about special agreements. Here it's the opposite.

Union agreements

Algorithms reflect the exact terms of your union agreements — including special clauses

Custom business rules

Unique pay rules, shift structures, or employee categories? They're built into the code, not worked around

Future changes

If agreements or rules change, the algorithms are updated — no need for a new system

Works with the systems you already have

The solution doesn't replace DK, H3, or any other payroll system. It acts as a bridge: takes data from whatever source you already have (shift files, Excel exports, databases) → performs the necessary calculations → returns an Excel file ready to import into your payroll system.

Shift files / Excel
↓ Python ↓
DK / H3 / other

Typical payroll process

  • Step 1:Program A - calculate day/night hours for shift workers
  • Step 2:Manually transcribe results to Excel
  • Step 3:Program B - enter data for regular employees (different interface)
  • Step 4:Manually verify each sick leave in separate system
  • Step 5:Program C - add meals, mileage, bonuses (another interface)
  • Step 6:Manually merge all data into payroll system

Validations: all by hand | Errors: frequent during transcription

With this solution

  • Step 1:Manager fills one Excel sheet with shifts — or uses an export from your existing scheduling system
  • Step 2:Upload file to system — click "Process"
  • Auto:System retrieves employment data from whatever source you already use (Azure, SQL, Excel)
  • Auto:Validates sick leave, classifies hours, calculates overtime and bonuses — according to your rules
  • Step 3:Download ready Excel — formatted exactly as DK, H3, or your payroll system requires

Validations: automated | Errors: zero | DK/H3: unchanged

The system adapts to you — not you to the system

Typical payroll systems

  • Fixed rules — the company must adapt its workflows
  • Special union agreements often don't fit the standard format
  • Changes require expensive custom development from the vendor
  • You pay for features you never use
  • Implementation takes months and can disrupt operations

This solution

  • Algorithms written specifically around your union agreements and pay rules
  • Unique shift structures or employee categories are implemented — not excluded
  • Rules change → code is updated. No one needs to buy a new system
  • You pay only for what you need — and only when it works
  • DK, H3, and your existing database remain untouched

Try Live Demo

Upload a sample shift file and see how the system automatically processes it — just like it would work alongside your existing DK or H3 setup.

Only .xlsx or .xls files

Download Sample File

Download a ready template with sample data to see the exact input structure. The output Excel will be formatted to match your payroll system's import requirements.

Download Excel Template

Template includes:

  • • Columns: Employee ID, start/stop date/time, shift type
  • • Sample data for 5 employees (shift and regular)
  • • Example codes: REG, VAC, SICK — but your company defines its own codes
  • • Correct date and time formatting

Shift codes defined by your company — no limit on the number

The system is not locked into a fixed set of codes. Each company defines its own shift codes and the system learns to classify and calculate based on them. Nobody needs to rename shifts in their existing systems.

Example: Hospital

  • DAY Day shift
  • NIGHT Night shift
  • ON_CALL On call
  • SICK_CERT Sick with certificate
  • TRAINING Training

Example: Restaurant

  • LUNCH Lunch service
  • DINNER Dinner service
  • EVENT Special event
  • PREP Preparation
  • HOL_PAY Holiday paid

Example: Factory

  • SHIFT_A Shift A (06-14)
  • SHIFT_B Shift B (14-22)
  • SHIFT_C Shift C (22-06)
  • SICK_1D Sick 1 day
  • OVERTIME Overtime

Unlimited number of codes — a single algorithm can classify and process many codes simultaneously

  • Codes can be in any language — Icelandic, English, numbers
  • New codes can be added without disrupting existing ones
  • Same shift can combine multiple codes — e.g. overtime on a holiday
  • If your existing scheduling system already uses its own codes, we only need to know what they mean

What the system does automatically

Each of these steps is designed around your rules — not a generic standard

Sick Leave Validation

For each employee, system:

  1. Retrieves employment date from your existing data source
  2. Identifies union agreement (determines entitled days)
  3. Searches all shifts in last 12 months
  4. Counts used sick days (separately own and children)
  5. If limit exceeded: automatically converts excess days to unpaid leave

Tailored: Sick day limits, calculation method, and handling of excess days are configured based on your specific union agreements

Work Hours Classification

For each shift, system:

  1. Retrieves individual employee settings (evening/night hours may differ)
  2. Checks if day is holiday (regular vs major holidays)
  3. Divides shift into periods: day, evening, night
  4. Calculates hours per each type (including holidays)

Tailored: Evening/night thresholds, holiday lists, and pay rate multipliers are configured to match your agreements

Overtime Calculation

System determines:

  1. Required work hours (working days × work ratio × month hours)
  2. Considers employment start date (if mid-period)
  3. Compares with actually worked hours
  4. Classifies overtime per union agreement (regular rate vs overtime)

Tailored: Overtime rate tiers, thresholds, and special rules for different employee groups are built in

Vacation and December Bonuses

Automatic calculation:

  1. Verifies if employee worked minimum 12 weeks
  2. Identifies shifts qualifying for bonus
  3. Calculates due bonus as of settlement date
  4. Includes in final payment (if employee leaving)

Tailored: Bonus eligibility conditions and calculation method follow your specific agreement

Additional Payments and Benefits

In the same Excel file you can specify other payments — system automatically includes them in the settlement, which is then ready to import into DK, H3, or any other payroll system:

Meal Allowances

With automatic tax information

Car Usage

Mileage reimbursement

Company Stock / Bonuses

One-time benefits

Python Notebooks: runs where you need it

The solution is built on Python notebooks — a flexible format that runs in virtually any Python-capable environment:

Local execution

  • Data never leaves the client's environment
  • Works without internet connection
  • No dedicated server setup required
  • Runs on any machine with Python

Cloud execution

  • Azure Function — automated scheduled execution
  • Microsoft Fabric — integrates with Power BI and Data Factory
  • Jupyter Hub, Databricks, or any other Python notebook platform
  • Easy updates and version control

Same code — different runtime environments. Your choice of platform doesn't affect the calculation results.

Integration with your current systems — without changing them

Data in (inputs)

  • Shift files from any scheduling system
  • Excel exports from your current systems
  • Databases (SQL Server, Azure SQL, etc.)
  • Manually filled Excel if no other source available

Data out (outputs)

  • Excel in DK import format
  • Excel in H3 import format
  • Custom format for any other payroll system
  • Modular — use only the parts you need (e.g. only sick leave validation)

No need to migrate data from existing databases or replace DK/H3. The solution works as a middle layer — the calculation logic sitting between your data source and your payroll system.

Want to see how it works alongside your existing setup?

Detailed presentation shows the full data flow — from shift files to DK/H3 import, step by step

Remember: you pay only when payroll automation succeeds. No risk.