Prototype. Nothing here uploads a file to a server. The demo below runs in your browser only, on file names you pick or type. It shows what the real service would return: a provenance log and a per-file loss report. Demo only. Nothing is sent, stored, or processed, and no real request was received. All sample files, tables, and row counts in the demo are fictional.
What is it? A one-job file converter with two candidates in one demo: a legacy archive batch converter, and an Access database to SQL, CSV, or XLSX migration. Both return a provenance log and a per-file or per-table loss report.
Who pays? For Access: small firms migrating off Access, bookkeepers, IT support, and data recovery. For legacy archives: law firms, archives, probate, and IT migration buyers. Both are buyers who pay for correctness, not for one generic file.
What problem does it solve? Old files are stuck in dead formats: WordPerfect, Lotus 1-2-3, Microsoft Works, Quattro Pro, Ami Pro, WordStar, old Access databases, and Outlook PST mailboxes. Free tools convert one file at a time and say nothing about what was lost. Desktop tools are Windows only, one-time, and give a demo preview only.
What would the customer receive? Modern files plus a per-file provenance log: source format, engine used, date converted, what was lost, and a checksum. For Access: clean tables as CSV, XLSX, or a SQL dump for PostgreSQL or MySQL, with per-table row counts. Forms, reports, and VBA macros are lost by design, and the report says so up front.
Why would they buy ours? The gap is batch plus provenance plus a SQL-dump target, in a cloud self-serve page, with corrupt-file handling. No cloud batch service with a provenance log and a per-file loss report was found in the research pass. That gap is narrow, and it only matters for corrupt-file and SQL-dump jobs, which the free and desktop tools do badly.
What would it cost? PROPOSED pricing, not validated. One-time and fixed-price is the model, not a subscription. Legacy archive: $49 per batch, up to 500 files, with a free 10-file sample. A monthly plan is only an alternative, not the default. Access: $29 per database, with a free preview of table names and row counts before paying. A monthly plan is only an alternative, not the default.
What evidence supports it? Access leg: 3 dated 2026 Upwork buyer job posts for Access migration, Sep 22 2026, Jul 13 2026, and Feb 28 2026, plus a Mar 11 2026 corrupt Access recovery ask. Buyers in those posts pay $2,500 fixed, $45 to $100 an hour, or $100 to $855 for a human service, not $29 for a self-serve tool. Legacy leg: only 1 old buyer ask, from 2008, for a batch WordPerfect conversion. That leg on its own fails.
What is still unproven? The $29 and $49 prices are untested. None of the buyer posts is for a $29 self-serve tool. There is also a scope mismatch: buyers in the Upwork posts want forms, reports, workflows, and VBA kept or recreated, while this prototype drops them by design. No search-volume numbers were verified. No buyer was found asking for a corrupt-file SQL dump at $29.
What happens next? The smallest test is a $0 landing page with a free table preview and a sample corrupt-file SQL dump, measuring paying databases in 30 days. Kill the Access leg if fewer than 20 paying databases in 30 days, or if fewer than 10 of those are corrupt-file or SQL-dump jobs. Kill the legacy leg if a free 10-file sample gets fewer than 20 genuine archive uploads in 30 days, or fewer than 5 paying batch customers. If the only demand is clean single-table CSV, the desktop $29 tools win.
Maturity, tracked separately:
Access buyer signals. Upwork, "Convert Access 2013 to Access 365", Sep 22 2026. Upwork, "System Programmer for MS Access to ERPNext Migration", Jul 13 2026, $2,500 fixed price. Upwork, "Microsoft Access Database Expert Needed for Migration", deadline Feb 28 2026, $45 to $100 an hour, buyer wants a move to SQL Server. Upwork prices buyers actually pay for Access migration services: $100 to $500, $190 to $855, $150, and $199 to $999. DBConvert for Access to SQL Server, $199, published 2026-11-08. Fixed-scope modernisation from $1,200.
Legacy buyer signals. Ask MetaFilter, "Archive WordPerfect?", May 24 2008, a batch of 10 to 100 WordPerfect files. That is 18 years old, and the answers point to free OpenOffice and a script. No recent dated buyer post was found asking for a paid $49 per batch legacy archive batch service.
Competitors. Legacy File Converter, $79.95 one-time, Windows desktop, no cloud, no batch API. OldFileOpener, free single file, no provenance log. WPD Converter, one-time from $39.95, 10 files free. PST tools, $29 to $49 one-time, desktop only. Access desktop tools: SysTools $69, Softaken $29, QuickData $69, DataHelp $69, all Windows desktop, one-time, demo previews only. No cloud API, no SQL-dump target, no batch migration service was found.
Sources. stream-file-converter-candidates.md, checked 2026-11-10. background-change-monitor-file-converter-demand-test.md, 2026-11-11, grade for this prototype: PASS, narrowly and conditionally, on the Access leg only. Key links: Upwork Access 2013 to 365, Upwork Access to ERPNext, $2,500, Legacy File Converter, Softaken $29.
Honest limitations. No search-volume numbers were verified in the research pass, and none are quoted here. The Access buyers in the posts want a working system with forms, reports, and VBA, which this tool does not deliver. The $29 price is far below what those buyers pay a human. Generic conversion is a bad bet: bank statements, STL repair, ebooks, and subtitles are free or well served, and Zamzar and CloudConvert cover the generic API job. Do not build a batch archive service on the strength of the Access job posts alone.
Generic file conversion is a bad bet in 2026. Bank statements, STL repair, ebooks, and subtitles are free or well served. These two candidates survive because they are not generic: batch, verification, a niche format, or a buyer who pays for correctness.
Upload a folder, a PST mailbox, or a whole archive drive. Out come modern files, plus a per-file provenance log: source format, engine used, date converted, what was lost, and a checksum. Batch, not one file at a time.
Formats in: WordPerfect (.wpd), Lotus 1-2-3 (.123, .wk1, .wk3, .wk4), Microsoft Works (.wps), Quattro Pro (.qpw), Ami Pro (.sam), WordStar (.ws), Access (.mdb, .accdb), Outlook PST (.pst).
Formats out: DOCX, XLSX, PDF, CSV, SQL, MBOX, EML.
Buyers: law firms, archives, probate, IT migration, companies leaving a dead system.
Price: PROPOSED, not validated. $49 per batch (up to 500 files), one-time. Free 10-file sample. A monthly plan is only an alternative, not the default.
$10k math: One-time batches do not recur. 204 batches at $49 in a month is $9,996 in that month, and that volume every month is not evidenced. A monthly plan is only an alternative, not the default, and it is not assumed here.
Kill condition: kill if a free 10-file sample gets fewer than 20 genuine archive uploads in 30 days, or fewer than 5 paying batch customers, or if no law-firm, archive, or migration buyer can be named with a dated public need.
Incumbents, checked 2026-11-10: Legacy File Converter, $79.95 one-time, Windows desktop, no cloud, no batch API. OldFileOpener, free single file, no provenance log. PST tools, $29 to $49 one-time, desktop only. Gap: no cloud batch service with a provenance log and a per-file loss report.
Upload a Microsoft Access database, including a corrupt one. Out come clean tables as CSV, XLSX, or a SQL dump for PostgreSQL or MySQL, with a per-table row count and a list of what was lost. Forms, reports, and VBA macros are lost by design, and the report says so up front. No Access licence needed.
Buyers: small firms migrating off Access, bookkeepers, IT support, data recovery.
Price: PROPOSED, not validated. $29 per database, one-time. Free preview of table names and row counts before paying. A monthly plan is only an alternative, not the default.
$10k math: 345 databases at $29 in a month is $10,005 in that month, and it does not recur. Repeat migration is rarer than a one-off move. That is the core weakness. A monthly plan is only an alternative, not the default, and it is not assumed here.
Kill condition: kill if fewer than 20 paying databases in 30 days, or if fewer than 10 of those are corrupt-file or SQL-dump jobs, the only jobs the free and desktop tools do badly. If the only demand is clean single-table CSV, the desktop $29 tools win.
Incumbents, checked 2026-11-10: SysTools $69, Softaken $29, QuickData $69, DataHelp $69, all Windows desktop, one-time, demo previews only. No cloud API, no SQL-dump target, no batch migration service found.
Pick a demo, add files or use the sample set, then run it. Demo only. Nothing is sent, stored, or processed. No real request was received, and no file is uploaded anywhere. For real local files, only the name and size are read, never the contents.
Sources and full research: stream-file-converter-candidates.md, checked 2026-11-10. No search-volume numbers were verified in that pass, and none are quoted here. That is itself the first cheap test for each candidate.