Frequently asked questions
Fit
- What is Feldspar?
- Feldspar is a company that builds software to turn messy inputs (documents, emails, attachments, photos, logs, spreadsheets) into clean, structured records and push them into the systems you already use: ERP, CRM, databases, or spreadsheets. We also add practical processing like quality checks, matching against existing records, and exception flags, so the output is directly useful.
- What problems does Feldspar solve?
- In many organisations, critical information is locked in documents, emails, and files that only a few people know how to process. That creates data silos, manual bottlenecks, and single points of failure: when that one person is unavailable, the process stalls. Feldspar automates these pipelines end to end. We extract the right fields, match them to your existing records, apply quality checks, and push the result into your systems. Less manual handling, fewer errors, and no more dependency on a single person to keep things moving.
- Who is Feldspar for?
- Feldspar is built for companies that deal with recurring volumes of messy, inconsistent documents, especially in industrial, logistics, and operational environments where manual data entry, copy-pasting, and quick-and-dirty fixes like Excel macros are slowing things down or introducing errors. It also fits subsidiaries, local entities, or departments within larger corporations that operate with the complexity of an SME but without the IT resources to match: real problems, limited appetite for heavy implementations.
- What are common starting points?
- Documents into systems of record: invoices, delivery notes, certificates, spec sheets, service reports, contracts
- Emails into structured cases: inbox messages turned into cases with the right fields, attachments bundled, duplicates removed
- Matching and cleanup: labels or free-text descriptions matched to the right database record, supplier names deduplicated, many formats merged into one consistent dataset
- Comparing, checking, and flagging: expenses checked against policy, version diffs, discrepancy flags and review queues
- Internal libraries: CVs turned into searchable profiles, proposals organised into reusable building blocks, project updates turned into structured client reporting
How it works
- Do we need a large IT project to get started?
- No. Feldspar is designed to be integrated with minimal effort: no lengthy onboarding, no complex configuration, and no need for dedicated internal resources.
- What do you need from us to start?
- Get in touch or try it right away: send a few sample files along with your requirements and we'll return a free preview of the desired output. To scope a full pipeline, we typically need:
- A set of representative sample files
- The fields you want extracted (e.g. supplier, date, reference, line items)
- Any additional processing: matching, checks, enrichments, or specific business rules you want applied
- What the output should look like (Excel/CSV/JSON/API or your database schema)
- What does the end-to-end flow look like?
- Feldspar takes data from your existing channels (upload, folder drop, email, or API), pulls the right fields out of documents and messages, maps them to your data model, standardises formats, and matches against existing records. On top of that, we add business-ready enrichment: deterministic checks, simple aggregations, derived fields, and exception flags. Anything uncertain goes through a light review flow. Outputs are then written back into your target systems (ERP, CRM, WMS, database, or API). Every pipeline includes monitoring, regression tests, and audit trails.
- Is this a product or a custom consultancy project?
- It's a product: software that runs continuously and reliably, not a one-off project. Feldspar keeps working without you owning scripts or babysitting brittle automation. We handle monitoring, regression tests, and updates so the pipeline stays accurate as your documents evolve.
Inputs & outputs
- What types of inputs does Feldspar handle?
- Documents, emails, attachments, free text, structured exports, database records, and API feeds. Think invoices, delivery notes, service reports, certificates, contracts, expense claims, equipment labels, machine logs, inbox messages with attachments, spreadsheet exports, and data pulled from existing databases or APIs. Both digital files and scanned or photographed inputs are supported.
- What input formats do you support?
- PDFs, images (photos/screenshots), scans, spreadsheets, text and binary logs, emails with attachments, and mixed bundles. If you're unsure, send a few samples. We'll tell you what's realistically extractable.
- What outputs do we get back?
- Clean, reliable records written back into the systems your team already uses: ERP, CRM, WMS, databases, or spreadsheets. Output formats include Excel, CSV, JSON, or direct API and database integration. On top of the structured data, we add business-ready enrichments: quality checks, record matching, derived fields, and exception flags, so the result is directly useful, not just structured.
- How does integration work with our current tools?
- Feldspar plugs into your existing setup. Inputs arrive from your real channels (upload, folder drop, email, or API), and outputs are written back into your target systems through an API or database integration. There is no need to change your current systems.
- Can Feldspar handle handwriting, low-quality scans, and non-standard layouts?
- Often yes, including common real-world issues like poor lighting, skewed scans, and inconsistent templates. That said, quality still matters: we'll be transparent about limitations after seeing samples.
Getting started
- How do we get started?
- Send a few sample files and we'll return structured output for free so you can see the quality first-hand. No commitment, no onboarding. If it looks good, we scope the smallest useful production pipeline together: the document types, fields, checks, and target system. No long program, just a clear path from preview to production.
Company
- Where is Feldspar based?
- Feldspar is based in Belgium, but we work with clients worldwide.
Still have questions? Let us show you with your own data.