
Welcome to DressenHeims Unternehmensberatung GmbH
Since 1997, we have been your partner for restructuring, performance improvement and the transformation of mid-sized trading companies. Today, we place particular emphasis on innovative AI projects.
Our expertise for your success
From analysing unstructured data to integrating artificial intelligence into your workflows, we deliver tailored solutions.
Restructuring
We help you realign your company and build a sustainable future.
Performance improvement
Performance improvement begins with an analysis of the business situation.
Transformation
Manage change in your company with our professional support.
AI projects
Harness the power of artificial intelligence to create innovative solutions for your business.
Our partners
Our partner for corporate finance services – IPONTIX Corporate Finance
IPONTIX Corporate Finance advises companies, many of them family-owned, on acquisitions and divestitures, succession solutions, strategic financing matters and company valuations. Its experienced team combines many years of corporate finance expertise with an extensive network of banks and investors.
Visit the IPONTIX websiteOur partner for AI hardware and software – BYTEC, supporting more than 2,000 resellers in Germany
BYTEC Bodry Technology GmbH is a German specialist IT distributor for modern IT infrastructure, data and storage solutions, networking, cloud technologies, IT security and artificial intelligence. BYTEC supports IT service providers with technical advice and project assistance.
Visit the BYTEC websiteAI-supported email processing – from preparation to a production-ready solution
The interactive calculator is loading. If this message remains visible on the published website, the embedded code could not be started.
Email Automation: Your Potential Check
DressenHeims Unternehmensberatung · Workshop assessment and scenario calculation
Record volumes, categories, identifiers and document structures. Complete the processing assumptions together during the workshop. This is not an AI test and does not guarantee an automation rate.
Email volume and period
Maximum 999,999 emails per period · 30 categories · 50 identifier types · 100 groups. Distinct master-data values and senders: up to 99,999. Layout/format variants and identifier lengths: up to 100. Test emails: up to 99,999. Percentages: 0–100%.
Open the categories, identifiers and results when you are ready to explore. Collapsing the tool preserves your inputs.
Where the browser supports direct folder selection, the save dialog opens in Documents. In Safari, the “File download location” setting determines whether one folder is used or Safari asks for every download.
Possible categories
categories. Enter a separate email volume for each category. The sum must match the total volume.
Each category has its own volume field below. The last number entered never replaces another category's volume.
Remove unwanted suggestions using “Remove”. Associated groups are removed as well. Existing inputs require confirmation. Other categories remain unchanged.
Identifiers and mandatory fields
For an identifier, the count means the number of distinct values in the relevant master data, such as 25,000 item numbers. The number of different identifier types is counted separately.
Always treat identifiers as text: a numeric customer number may contain leading zeros. A unique match may also require a combination, such as manufacturer + item number or customer + order number.
Document structure and processing groups
Record only the emails not yet automated and their data sources here. A single group inherits the remaining volume from step 1: category volume × (100% − current automation rate). For multiple groups, allocate that remaining volume. Already automated emails are excluded. After changing the current share, multiple groups may need to be reallocated.
Today, future and additional potential
How is the estimate calculated?
For each group: not-yet-automated email volume × confident correct recognition × complete extraction × successful business validation × successful execution. Each rate after recognition applies to cases that passed the preceding stage. The calculation does not assume independent stages.
Already automated today = category volume × your entered current automation share. Additional potential applies only to the remaining volume. Future = already automated volume + additional potential. Existing automation is never double counted.
The source in step 1 describes existing automation. Sources, structures and groups in step 3 describe only emails not yet automated. Unknown current shares remain unknown; explicitly enter 0% if nothing is automated yet.
Presets: recognition 80–99% for category 1, 75–94% for category 2, and 70–89% from category 3 onward. The other stages use 90–100%, 85–90% and 100–100% for all categories. These editable values are orientation assumptions, not industry benchmarks or measured results. More reliable rates require detailed process analysis in the second company workshop and business-reviewed tests.
The range combines conservative and optimistic inputs. It is not a guarantee or a statistical confidence interval. Custom values are preserved; deliberately cleared fields remain unknown. The future estimate assumes an available interface but never bypasses mandatory human approval.
Additional workshop data
For each process, record other mandatory values: quantity and unit, price and currency, discount, validity, delivery date and address, payment terms, tax code and version. Also check for multiple line items, amendments in email threads, conflicting attachments, duplicates, password-protected documents and exception/approval rules.
For a proof of concept, provide labelled examples, exact field mappings, ERP interface documentation, reference data and named business reviewers. Evaluate recognition using precision and recall, and also measure fully correct transactions across the entire process.