MapFlow
The Hard-Hitting Case for MapFlow

Stop Migrating Data the Hard Way

Data migrations consistently overrun and fail for the same preventable reasons. MapFlow exists to eliminate them — drastically reducing risk, slashing effort, and collapsing timescales.

Spreadsheets get overwritten. Tribal knowledge walks out the door. PII is discovered in production. Validation rules reject rows mid-load. MapFlow fixes every one of these — by design.

1

95%+

Field-mapping confidence with candidate records

Versus 70-80% name-matching fallback

2

Weeks to Minutes

Screen Grab to Mapper pipeline

AI-driven discovery vs manual inspection

3

100%

Traceable, versioned, signed-off

Every change recorded forever

4

0 code

Required to ship a complete SQL view

No developer, no spreadsheets

5

up to 90%

Salesforce Storage Tax eliminated

Just shrink files via ConvertAPI before hitting Salesforce

6

24/7

AI bot auto-replies in Slack & Teams

@Mapflow mentions answered instantly — no context switching

Three Reasons MapFlow Wins

Every feature ladders up to one of these three outcomes.

Drastically Reduces Risk

Catch issues at design time, not production

MapFlow surfaces PII, quality issues, mapping gaps, and Salesforce validation conflicts before a single record moves. Traditional tools discover these during or after the load — when fixes are 10x more expensive.

  • AI PII scanning on schema and row data — GDPR Article 25 by design
  • Validation AI checks every mapping against live SF metadata
  • Full audit trail — who approved what, when, and why
  • Risk & issue log linked to every mapping and table
  • Snapshot-backed SF metadata toggle — org never left in a disabled state

Reduction

90%

Issues found at go-live to Issues found at design time

Slashes Cost

ConvertAPI kills the Salesforce Storage Tax

~85% of council documents are case files that must land in Salesforce at ~£9.90/GB/month. ConvertAPI compresses them up to 90% before upload, and a SharePoint hybrid route handles the subset that can live outside Salesforce. The storage savings alone offset the entire MapFlow licence.

  • ConvertAPI compresses PDFs up to 90% and images 70%+ before Salesforce
  • Case files shrunk at the source — tax eliminated
  • SharePoint hybrid route: ~£0.16/GB vs Salesforce ~£9.90/GB
  • Client-supplied licence, UK/EU data residency, GDPR-compliant
  • Storage savings offset the MapFlow licence — £0 net cost

Reduction

~£85k/yr

Full storage tax to £0 net cost

Collapses Timescales

From discovery to sign-off in a fraction of the time

Cloud execution, parallel workers, and AI-assisted discovery collapse the migration timeline. The overnight pipeline runs end-to-end without a consultant in the seat. Sign-off happens in-platform, not over email.

  • Schema imported from live DB in seconds, not days
  • AI generates UAT test plans instantly
  • Cloud parallel processing — no 2 GB Access ceiling
  • Automated schema re-sync handles late changes without restart
  • Built-in training environment onboards new consultants fast

Reduction

70%

Months of manual work to Weeks of guided work

Key Features — The Hard-Hitting Case

Why you should do it. The value. The risk reduction. The effort and timescale impact.

AI Mapping Assistant (NLP)

Why You Should Do It

Field mapping documents live in disconnected spreadsheets no one keeps up to date. Developers build from verbal briefs. Specs are ambiguous and drift.

The Value

Describe what you want — 'Map Customers to Account' — and the AI produces confidence-scored mappings with PII flags, code/value maps, and transformations.

Risk

Removes ambiguity. Every mapping has a written AI reasoning trail. No more tribal knowledge walking out the door.

Effort

Weeks of manual column-by-column inspection reduced to a conversation.

Timescale

Minutes to first draft mapping — versus days in spreadsheets.

⭐ Reverse-Engineering with Candidate Records

Why You Should Do It

Name-matching is unreliable. Cryptic column names (DT_ID, PL_REF_NO) defeat manual inspection. You're guessing.

The Value

Supply matched sourcetodestination record pairs — actual data values. The AI compares values side-by-side for 95%+ confidence. The 'follow me' approach: show the AI how to map by example.

Risk

Definitive evidence — when source APP_NO matches destination Application_Number__c value-for-value, that IS the mapping. No guesswork.

Effort

3-5 candidate records replace hours of cross-referencing documentation.

Timescale

95%+ confidence in one pass — versus 70-80% with multiple revision cycles.

Screen Grab to Mapper Pipeline

Why You Should Do It

Manually transcribing field labels from UI screenshots into a spreadsheet is slow, error-prone, and loses visual context.

The Value

Capture form fields and screenshots via bookmarklet, cross-reference with live DB columns, send the complete package to the AI. Mappings generate automatically with numbered markers on screenshots.

Risk

No transcription errors. Visual context preserved. Every field linked to its screen origin.

Effort

Eliminates manual screen-to-spreadsheet transcription entirely.

Timescale

Weeks of manual UI discovery compressed to minutes.

Postgres Staging Architecture (ELT)

Why You Should Do It

Running heavy joins and transformations against a live production source database risks locking tables and degrading performance for real users.

The Value

Source is copied to a dedicated, independently-sized Postgres database. All heavy work runs there. The source is read-only and lightweight — never asked to join millions of rows.

Risk

A migration run can NEVER degrade the live business system. The safe-to-run guarantee.

Effort

Environment-tiered schemas (dev/test/live) eliminate environment juggling.

Timescale

Postgres scales horizontally — no 2 GB Access ceiling, no manual batch splitting.

Unattended Overnight Pipeline

Why You Should Do It

Consultants babysitting browser tabs for hours is expensive and fragile. Manual metadata toggling in SF Setup is error-prone.

The Value

Three phases run while you sleep: source data refresh to Postgres, scheduled SQL transformations, and unattended Postgres to Salesforce loading. SF metadata auto-disabled and restored. Failure review and admin alerts built in.

Risk

SF metadata always restored — even on failure. Per-batch logs with per-record error drill-down. Admin email on every failure.

Effort

Zero consultant hours during the load. No browser tab, no babysitting.

Timescale

~12,000 rows/hour, runs for as many hours as needed across short calls.

Post-Go-Live Delta Sync & Reconciliation

Why You Should Do It

After go-live, legacy and Salesforce run in parallel. Without delta sync, every night is a full table reload. Without reconciliation, silent Salesforce edits get overwritten with no detection.

The Value

Watermark-based delta loading fetches only modified rows. Dual-side reconciliation compares source changes against Salesforce SystemModstamp changes — classifying source-only, SF-only, and conflicts. Reports saved to Document Library and emailed.

Risk

Detects when a Salesforce edit has been silently overwritten. Source is master; conflicts flagged for review. Full audit trail.

Effort

No full table reload each night — only changed rows processed.

Timescale

Overnight delta sync keeps systems aligned through the dual-running transition.

Validation AI Assistant

Why You Should Do It

Validation rule failures, required-field gaps, and picklist mismatches are discovered during the load — when fixes are expensive and block the cutover.

The Value

Automated agent checks every field mapping against live Salesforce metadata — required fields, picklist constraints, data types, and validation rules — at design time.

Risk

Issues fixed at design time are 10x cheaper than fixes during or after the load. Accept/reject decisions persisted per issue.

Effort

No manual SF Setup inspection. No 'did we miss a required field?' panic at go-live.

Timescale

Design-time validation runs in seconds — versus hours of manual review.

Governance, Sign-Off & Versioning

Why You Should Do It

No clear audit trail of who approved what and when. When a consultant leaves, institutional knowledge leaves with them. Spreadsheets get overwritten with no recovery. Risks and issues tracked in scattered lists with no visual workflow.

The Value

Semantic versioning with restorable snapshots. Three-role sign-off workflow (Business, Consultant, PM). Merge review with per-diff accept/decline. Comprehensive Client Management Report PDF per mapping. Trello-style Kanban board for drag-and-drop tracking of risks, issues, tasks, and to-dos across open, in-progress, resolved, and closed columns.

Risk

Every decision recorded forever. Every version restorable. The mapping artefact outlives any tool or any consultant. Visual Kanban board ensures no risk or task falls through the cracks.

Effort

Sign-off happens in-platform — no email chains, no verbal confirmations. Drag-and-drop status updates replace manual list maintenance.

Timescale

Board-ready PDF report generated in one click — versus days of manual document assembly.

Security & Compliance Built-In

Why You Should Do It

PII discovered in production rather than during design. No MFA. No session timeout. No encryption. No backup. Spreadsheets emailed around on shared drives.

The Value

TOTP-based MFA, automatic session timeout, RBAC with row-level security, full audit trail, AI PII scanning at design time, AES-256-GCM encryption for Postgres PII fields, and one-click backup & restore.

Risk

NIST SP 800-63B AAL2, ISO 27001, GDPR Articles 25 & 30, SOC 2, PCI DSS v4.0 — mapped and ready for audit.

Effort

Security is configured, not bolted on. No separate compliance project.

Timescale

Audit-ready from day one — versus months of remediation post-incident.

CDC Diff Reports & Document Library

Why You Should Do It

Every sync is a full wipe-and-replace with no visibility into what actually changed. Reports scattered across local machines.

The Value

Snapshot differencing detects row-level inserts, updates, and deletes. Central Document Library stores every report — CDC diffs, management reports, audit reports — searchable from any machine.

Risk

Know exactly what changed between syncs. Migrate only changed rows instead of full tables.

Effort

No ad-hoc report files on local machines — everything centralised and searchable.

Timescale

Pre-cutover CDC diff identifies changed records — migrate deltas, not the whole table.

Hybrid SharePoint / Arcus Document Loading

Why You Should Do It

Most documents must land in Salesforce ContentVersion so Arcus can see them — but a subset can be stored more cheaply in SharePoint and linked via Public_Share__c. Without a hybrid route, every document pays the Salesforce Storage Tax.

The Value

For documents that can live outside Salesforce, a per-upload hybrid route stores them in SharePoint (~£0.16/GB/month) and links to Arcus via the Public_Share__c + ContentDocumentLink bridge — ~98% cheaper than Salesforce file storage. A copy_flag column decides per document whether it copies to Salesforce (truthy) or links to SharePoint only (falsy).

Risk

Documents stored in SharePoint are recoverable and version-controlled. Public_Share__c provides the Arcus link without duplicating file storage in Salesforce.

Effort

No SFTP, no manual metadata re-entry. SharePoint upload and Arcus linking happen in one flow. Per-upload choice — no reconfiguration between uploads.

Timescale

Per-upload decision — hybrid or direct mode selected on each batch.

⭐ ConvertAPI Compression — Killing the Salesforce Storage Tax (MapFlow Pays for Itself)

Why You Should Do It

Roughly 85% of council documents are case files — high-fidelity masters (scanned PDFs, high-res photos). These 'fat' files must land in Salesforce ContentVersion so Arcus can see them. At ~£9.90/GB/month, every uncompressed upload is a recurring monthly charge — forever. Across terabytes, this becomes a persistent, compounding levy: the Salesforce Storage Tax.

The Value

ConvertAPI compresses those unredacted masters by up to 90% (PDFs) and 70%+ (images) BEFORE they reach Salesforce ContentVersion. Each client supplies their own ConvertAPI licence; the secret is stored per-project and used server-side only. Processing is in-memory (zero persistence) with UK/EU data residency — GDPR-compliant. At 1 TB of documents, a blended 70% compression removes ~0.7 TB from Salesforce — a saving of ~£7,100/month, or ~£85,000/year. The savings alone offset the entire MapFlow licence — £0 net cost.

Risk

Compressed documents are easier to preview, faster to load, and never breach Salesforce file-size limits. No file is persisted on ConvertAPI's servers.

Effort

Compression runs automatically inside the bulk loader — no manual pre-processing step, no extra tool to babysit.

Timescale

Per-document compression adds <2s to each upload; the parallel worker pool hides the latency across thousands of files.

Slack & Microsoft Teams Bot Integration

Why You Should Do It

Migration questions and issues get lost in email chains. Consultants and clients need instant answers without leaving their collaboration platform. Support ticket notifications go unseen.

The Value

AI-powered bot auto-responds to @Mapflow mentions in Slack channels and direct messages via real-time webhooks. Microsoft Teams bot polls channels for Mapflow mentions and generates AI replies. Support ticket notifications and connector onboarding conversations are routed directly to Slack and Teams for real-time visibility.

Risk

No missed support requests — every ticket notifies the team channel instantly. Connector credentials purged on conversation close.

Effort

Answers to common questions delivered in-channel by AI — no context switching to email or a separate portal.

Timescale

Real-time Slack webhook responses (instant). Teams polls every 5 minutes. Support tickets created and notified in one click.

Support Tickets & Connector Onboarding

Why You Should Do It

Support requests are scattered across emails, phone calls, and verbal handovers. Connector setup credentials are shared insecurely via email or chat. No audit trail of who requested what.

The Value

In-app support ticket system with structured categories, priority levels, and real-time Slack notifications. Connector onboarding conversations with secure credential exchange and automatic purging on close. Admin-managed with full audit trail and message history.

Risk

Every support request tracked, categorised, and auditable. Credentials purged when conversations close — no lingering secrets.

Effort

One-click ticket creation from anywhere in the app. No email chains, no lost requests.

Timescale

Ticket created and Slack notification sent in under 2 seconds. Connector conversations close with one click.

AI Bot Built Into Slack & Microsoft Teams

MapFlow isn't just a web app — it lives where your team already works. Mention @Mapflow in any Slack channel or direct message and the AI bot responds instantly with migration guidance, field-mapping advice, or platform help. Microsoft Teams integration polls channels for Mapflow mentions and replies with AI-generated answers. Support ticket notifications and connector onboarding conversations are routed directly to Slack and Teams — no one misses a request.

Instant

Slack webhook responses

5 min

Teams polling interval

24/7

AI auto-replies — no human needed

2 platforms

Slack & Microsoft Teams

Legacy migration tools have no collaboration platform integration whatsoever — no bot, no AI, no real-time notifications. Their support model is email and phone. MapFlow brings the answers to where your consultants and clients already are.

Storage Economics

Eliminating the Salesforce Storage Tax

At ~£9.90/GB/month, Salesforce file storage is billed at premium rates. Every uncompressed document you upload is a recurring monthly charge — for the lifetime of the system. Across terabytes of council documents, this becomes a persistent, compounding levy: the Salesforce Storage Tax. MapFlow kills it at the source.

Why the Tax Is So High: ~85% Case Files

The dominant payload driving the Storage Tax.

Roughly 85% of council documents are case files — high-fidelity masters with zero black bars or hidden text. These "fat" files (scanned PDFs at 20 MB, high-resolution photos at 10 MB) are the bulk of the library, and they must land in Salesforce ContentVersion so Arcus can see them. Without compression, they inflate the Storage Tax every month, forever.

~85%

Documents are case files

~£9.90

Per GB / month in Salesforce

Terabytes

Typical council document volume

Forever

Billed monthly, compounding

The Tax Exemption

Local Proxy Compression — Before Salesforce

The MapFlow Local Proxy compresses those case files by up to 90% (PDFs) and 70%+ (images) before they reach Salesforce ContentVersion. The open-source toolchain (pdfsizeopt, Ghostscript, jpegoptim, optipng) runs entirely on the council's own proxy host — no third-party sub-processor, no cross-border processing, no file content leaving the network. Lossless by default for a paid reduction service.

  • PDFs compressed up to 90%, images up to 70%+ before upload
  • Open-source toolchain on the council's own proxy host — no sub-processor
  • Document bytes never leave the network perimeter — strongest InfoSec posture
  • Lossless by default — preserves master/legal quality for a paid service
  • Runs automatically inside the bulk loader — no manual pre-processing

At 1 TB of documents

A blended 70% compression removes ~0.7 TB from Salesforce — a saving of ~£7,100/month, or ~£85,000/year.

Secondary Route

SharePoint Hybrid — For the Subset

Not every document must live in Salesforce. For the subset that can be stored more cheaply, a per-upload hybrid route stores them in SharePoint (~£0.16/GB/month) and links to Arcus via the Public_Share__c + ContentDocumentLink bridge — ~98% cheaper than Salesforce file storage. A copy_flag column decides per document whether it copies to Salesforce (truthy) or links to SharePoint only (falsy).

62×

Cheaper than Salesforce file storage

Compare SharePoint vs Salesforce storage costs for any volume — and see the monthly savings MapFlow unlocks.

Storage Cost Comparison

SharePoint vs Salesforce for bulk document storage

SharePoint

£162/month

Additional storage at ~£0.16/GB/month on top of M365 licensing

1,024 GB at selected tier

Salesforce

£10,138/month

File Storage add-on at £99/month per 10 GB

1,024 GB at selected tier

Published list rate — Arcus may have negotiated a reduced rate

Saving £9,976/month by using SharePoint — a 63× cost reduction

Hybrid Scenario — what % of documents need to be in Salesforce?

% Public (→ SF)SalesforceSharePointMonthly TotalAnnual SavingReduction
10%£1,014£146£1,160£107,73689%
20%£2,028£130£2,158£95,76079%
30%£3,041£113£3,154£83,80869%
40%£4,055£97£4,152£71,83259%
50%£5,069£81£5,150£59,85649%

Based on 1 TB total. The remainder stays in SharePoint at ~£162/TB/month; only the public percentage is copied to Salesforce ContentVersion at ~£10,138/TB/month. Pick the row that matches your project's document mix.

Key Considerations
  • SharePoint pricing assumes Microsoft 365 is already licensed — only additional storage is costed.
  • Salesforce File Storage add-on is £99/month per 10 GB beyond the 2 GB/user included allocation.
  • Salesforce figures shown are the published list rate — Arcus councils may have negotiated a reduced rate, so actual costs could be lower.
  • Strategy: keep non-public documents in SharePoint (low cost); sync only public-facing or Arcus-required documents to Salesforce ContentVersion.
  • Salesforce also counts file storage against your org's data allocation — large files can push you into higher edition requirements.

The Net-Zero Outcome — MapFlow Pays for Itself

The storage savings alone — Local Proxy compression on the ~85% case files, plus the SharePoint hybrid route for the subset that can use it — exceed the entire MapFlow licence fee. No other feature needs to be used. This single economics change makes MapFlow a £0 net-cost option for the project. Everything else — AI mapping, governance, overnight automation, dual-running reconciliation — is effectively free.

up to 90%

Smaller files before Salesforce

~£85k/yr

Storage Tax avoided (per 1 TB)

£0

Net cost to the project

The Cherry on Top

Arcus Market-Specific Tools

Generic platforms like Informatica give you extraction and field mapping. They don't give you the market-specific tools that councils migrating to Arcus actually need — the processing tools that handle documents, notes, images, and legacy application screens. MapFlow does.

IDOX Uniform Tools

A suite of tools purpose-built for councils migrating from IDOX Uniform — schema discovery from live systems, PDF documentation parsing, and Uniface metadata extraction.

  • Schema Scraper — extracts table & column definitions directly from the IDOX Uniform web interface
  • PDF Schema Scraper — parses IDOX database documentation PDFs to build a complete schema catalogue
  • Uniface Parser — reverse-engineers Uniface form definitions to discover non-DB fields and relationships
  • IDOX Migration Proposal — generates a structured migration plan from discovered schema metadata

Why generic tools can't: No generic tool understands IDOX Uniform's schema conventions, Uniface form layers, or IDOX PDF documentation format. These tools do.

SQL Server Utils

Licenced scripts that run on-premise against live SQL Server databases — extracting binary file content and RTF data without exposing credentials outside the client network.

  • Binary File Exporter — extracts BLOB documents (PDFs, TIFFs, images) from SQL Server to the local filesystem
  • RTF Extractor — identifies and extracts RTF column data to SQLite, Excel, MS Access, or CSV targets
  • Credentials never embedded in scripts — fetched at runtime securely

Why generic tools can't: Informatica doesn't ship a tool that runs on a council's network and extracts binary document blobs. This does.

Tiff Flow

Specialised TIFF image processing pipeline for planning and building-control document migrations — handles multi-page TIFFs, page splitting, and format conversion as part of the document load.

  • Multi-page TIFF splitting — separates bundled document scans into individual records
  • Format conversion — TIFF to PDF/JPEG for Salesforce ContentVersion compatibility
  • Metadata preservation — retains document indexing and folder structure through conversion
  • Batch processing — handles tens of thousands of images overnight without manual intervention

Why generic tools can't: Generic ETL tools treat TIFFs as opaque blobs. Tiff Flow understands multi-page planning application scans and splits them correctly.

Document & Note Loaders

Purpose-built Salesforce loaders for documents, notes, and attachments — with bulk processing, relinking, and per-record error tracking. Not generic data loaders; these understand Salesforce ContentVersion and Note objects.

  • Document Loader — uploads files to Salesforce ContentVersion with field-mapped metadata
  • Note Loader — creates Salesforce Notes with formatted body text from source records
  • Bulk Document Loader — parallel processing of thousands of documents with restartability
  • Document Relink — re-associates documents to corrected parent records post-migration
  • Per-batch and per-record error logging — drill down to the exact row that failed

Why generic tools can't: Informatica loads rows. These tools understand Salesforce's ContentVersion, Note, and Attachment object model — including the quirks of file vs. metadata loading.

PDF Merger

Combines multiple source documents (letters, plans, certificates) into single Salesforce ContentVersion records — the way case officers expect to see them, not as scattered individual files.

  • Rule-based merging — combine documents by application reference, document type, or date
  • Cover page generation — auto-creates index pages for merged document sets
  • Bookmark creation — navigable table of contents within merged PDFs
  • Source tracking — every merged file retains a link to its original source record

Why generic tools can't: No generic tool merges planning documents by application reference with auto-generated cover pages and source tracking. This is a migration-specific capability.

Document Audit

Post-load verification tool that compares documents loaded into Salesforce against the source system — checking file counts, sizes, types, and content integrity. Catches silent drops, corruption, and mismatches before the client notices.

  • File-by-file comparison — source vs Salesforce ContentVersion, with size and type validation
  • Batch audit tool — runs offline on exported source data with a licensed script
  • Path substitution & path repair — fixes broken document links after folder restructure
  • Audit warning badges — flags size mismatches, type mismatches, missing files, and manual overrides
  • Sign-off summary — found / missing / unchecked / errors so the client can see the audit state at a glance

Why generic tools can't: Informatica loads rows and moves on. Document Audit verifies every single document arrived intact — the difference between 'we loaded it' and 'we can prove we loaded it correctly'.

Screen Grabber

Captures legacy application screens (IDOX Uniform, Academy, etc.) and uses AI vision to extract field labels, positions, and relationships — cross-referencing them with live database columns for instant field mapping.

  • AI vision field extraction — detects form fields, labels, and layout from screenshots
  • Numbered marker overlay — visual trace from screen position to database column
  • Candidate record capture — 'follow me' approach: show the AI how to map by example
  • Session save & resume — persist screen grab sessions for team review

Why generic tools can't: Informatica can't see your screen. The Screen Grabber turns visual UI context into structured mapping data — weeks of manual transcription eliminated.

These tools ship with MapFlow — not as add-ons, not as partner integrations.

A client buying Informatica still needs someone to write the document loader, build the TIFF splitter, reverse-engineer the IDOX schema, and capture the legacy screens. With MapFlow, that's already built, tested, and integrated into the migration pipeline.

Structurally Impossible

What Datachecker & Spreadsheets Can Never Do

These aren't missing features on a roadmap. They are technical impossibilitiesfor tools built on a ~30-year-old Access + Excel VBA stack. No amount of patches, plugins, or consultant hours will close these gaps.

The 30-Year-Old Problem

Arcus Spreadsheet runs on Excel VBA. MS Access Datachecker is a compiled Access executable. Both predate the cloud, predate modern AI, predate web browsers as a platform, and predate the data-protection regulations your project now must satisfy. They cannot be "upgraded" into the modern era — their architecture is the ceiling.

AI field mapping & natural language

Legacy Cannot

No AI, no NLP. ~30-year-old VBA stack has no mechanism for AI integration. Mappings entered cell-by-cell.

MapFlow Does

AI Mapping Assistant produces confidence-scored mappings from a sentence. Reverse-engineering hits 95%+ confidence.

Why it's impossible: AI requires a modern runtime, API access, and a data model — none of which exist in Excel VBA or compiled Access.

Reverse-engineering with candidate records

Legacy Cannot

No value-based matching. Consultants eyeball column names and guess — typically 70-80% confidence at best.

MapFlow Does

Supply matched sourcetodestination record pairs. AI compares actual values side-by-side for definitive 95%+ mappings.

Why it's impossible: Requires a comparison engine, schema introspection, and LLM reasoning — structurally absent from spreadsheet tools.

Cloud-scale data processing (>2 GB)

Legacy Cannot

MS Access has a ~2 GB practical ceiling. Large datasets require manual batch splitting and Compact & Repair.

MapFlow Does

Postgres staging scales horizontally. No ceiling — millions of rows processed in parallel by cloud workers.

Why it's impossible: Access is a single-file desktop database engine. It cannot be horizontally scaled — that's a fundamental architectural limit.

AI PII scanning at design time

Legacy Cannot

No PII detection whatsoever. PII is discovered in production — a GDPR Article 25 violation waiting to happen.

MapFlow Does

AI scans schema metadata and Local Table row data before data moves. PII flagged, encrypted, or obfuscated at the source.

Why it's impossible: PII detection needs an LLM and a data-sampling pipeline. Spreadsheets have neither — and cannot acquire them.

Multi-factor authentication (MFA)

Legacy Cannot

Windows/network credentials only. No TOTP, no per-user enforcement, no backup codes, no lockout policy.

MapFlow Does

TOTP-based MFA, admin-enforced per user, 24-hour lockout, backup codes. NIST SP 800-63B AAL2 compliant.

Why it's impossible: MFA requires an identity provider and token verification — Excel and Access have no identity layer.

Automatic session timeout

Legacy Cannot

Spreadsheets stay open indefinitely. Access apps stay open until manually closed. No inactivity detection.

MapFlow Does

Configurable 10 min–2 hr timeout tracking mouse, keyboard, scroll, and touch. Global, immediate.

Why it's impossible: Session management is a server-side concept. Desktop files have no session lifecycle to terminate.

Real-time multi-user collaboration

Legacy Cannot

Shared-drive spreadsheets cause merge conflicts. Datachecker is single-user desktop only.

MapFlow Does

Cloud-based. Multiple consultants work in the same project simultaneously with row-level security.

Why it's impossible: Real-time collaboration needs a centralised database with conflict resolution — spreadsheets are file-based by nature.

Version control & restorable snapshots

Legacy Cannot

Files get overwritten. No change history. No way to recover a previous state. Knowledge leaves with the consultant.

MapFlow Does

Semantic versioned snapshots. Restore any previous version. Full audit trail of who changed what and when.

Why it's impossible: Versioning needs a database backend with append-only history. Files on a shared drive have no such capability.

Unattended overnight automation

Legacy Cannot

Every step requires a consultant at a keyboard. No scheduling, no unattended execution, no failure alerts.

MapFlow Does

End-to-end overnight pipeline: refresh to transform to load. SF metadata auto-toggled. Admin email on failure.

Why it's impossible: Unattended execution needs a server runtime, a scheduler, and API orchestration. Desktop tools cannot run headless.

Automated SF metadata toggle (validation/triggers/flows)

Legacy Cannot

Validation rules and triggers must be disabled manually in SF Setup before each load, then re-enabled by hand. Error-prone.

MapFlow Does

Auto-disables and restores validation rules, Apex triggers, and flows via the Tooling API. Snapshot-backed. Always restores.

Why it's impossible: Requires Salesforce Tooling API integration and a server to orchestrate it. Excel and Access have no API integration layer.

Post-Go-Live delta sync & reconciliation

Legacy Cannot

Every load is a full wipe-and-replace. No delta awareness. No change detection. Silent Salesforce edits overwritten with no record.

MapFlow Does

Watermark-based delta loading. Dual-side reconciliation via SystemModstamp. Conflicts flagged. Reports to Document Library.

Why it's impossible: Delta sync needs a persistent watermark store, a scheduler, and a comparison engine — none exist in legacy tools.

Validation AI against live Salesforce metadata

Legacy Cannot

Validation rules and triggers not surfaced by the Updater. Must be reviewed manually in SF Setup. Discovered during the load.

MapFlow Does

Automated agent checks every mapping against live SF metadata at design time — required fields, picklists, types, rules.

Why it's impossible: Live metadata introspection + AI reasoning against it requires an API client and an LLM — absent in desktop tools.

Slack & Microsoft Teams bot with AI auto-replies

Legacy Cannot

No collaboration platform integration. Support requests via email and phone. No real-time notifications. No AI bot.

MapFlow Does

AI-powered bot auto-responds to @Mapflow mentions in Slack channels and DMs via real-time webhooks. Teams bot polls and replies. Support ticket notifications routed to Slack.

Why it's impossible: Requires an LLM, OAuth connector infrastructure, and webhook receivers — none of which exist in spreadsheet or desktop database tools.

In-app support tickets with secure connector onboarding

Legacy Cannot

Support requests scattered across emails and calls. Connector credentials shared insecurely. No audit trail.

MapFlow Does

Structured support ticket system with categories, priority, Slack notifications, and admin chat. Connector onboarding conversations with credential purging on close.

Why it's impossible: Requires a database backend, authentication, and structured workflow — absent from file-based legacy tools.

Pre-upload document compression (ConvertAPI) — kills the Salesforce Storage Tax

Legacy Cannot

Documents are uploaded at their original size. ~85% are unredacted masters — scanned PDFs at 20 MB, high-res photos at 10 MB. Each costs ~£9.90/GB/month forever in Salesforce. Across 100,000 documents that's thousands of pounds per month of avoidable Storage Tax.

MapFlow Does

ConvertAPI compresses unredacted masters by up to 90% before they reach Salesforce — eliminating the Storage Tax at the source. UK/EU data residency, in-memory, GDPR-compliant. Client provides the licence.

Why it's impossible: Requires an API integration, a per-project secret store, and a server-side compression pipeline — none of which exist in spreadsheet or desktop database tools.

You cannot retrofit AI into a spreadsheet. You cannot scale Access to the cloud. You cannot bolt governance onto a shared drive.

These aren't product gaps to be logged in a backlog — they are the structural ceiling of the legacy stack. MapFlow was built ground-up on a modern, cloud-native, AI-first architecture. That's why it can do what they cannot.

The ROI Gap

Traditional approaches vs MapFlow — metric by metric.

Metric
Traditional
MapFlow
Time to first mapping draft
1-2 weeks
Minutes
Mapping confidence (with candidate records)
70-80%
95%+
Issue discovery point
During/after load
Design time
Overnight load consultant hours
8+ hours babysitting
0 hours (unattended)
Audit trail & version history
None — files overwritten
100% — every change recorded
PII detection timing
In production
At design time
Dataset size ceiling
~2 GB (MS Access)
Unlimited (cloud)
SF metadata toggle
Manual, hours, error-prone
Automated, snapshot-backed
Salesforce Storage Tax (document storage)
Premium per-GB/month — billed in full, forever
ConvertAPI kills it — up to 90% smaller files
Document file size before Salesforce
Original size — 85% are fat unredacted masters
Compressed up to 90% (ConvertAPI)
Collaboration platform bot
None — email and phone only
Slack & Teams AI auto-replies
Risk & task tracking
Scattered spreadsheets & email
Visual Kanban board — drag & drop

Risks We Eliminate

Every row is a migration failure mode MapFlow makes impossible.

Mapping drift between versions
Eliminated

Versioned snapshots, merge review, full audit trail

PII leaked into Salesforce
Eliminated

AI PII scanning at design time + AES-256-GCM encryption

Production source degraded by migration
Eliminated

Postgres staging — source is read-only and lightweight

Go-live cutover failures
Eliminated

Validation AI + pre-load metadata toggle + design-time checks

Consultant knowledge loss
Eliminated

Every mapping documented, versioned, and signed off in-platform

Silent Salesforce edits overwritten
Eliminated

Post-Go-Live reconciliation detects and flags dual-side changes

SF org left in disabled state
Eliminated

Metadata restore runs in the catch block — always restores

Salesforce Storage Tax on unredacted masters
Eliminated

ConvertAPI compresses the ~85% unredacted masters up to 90% before Salesforce upload — tax eliminated at the source

No recovery from bad changes
Eliminated

One-click backup & restore + versioned snapshots

The Cost of Doing Nothing

Spreadsheets get overwritten with no recovery. Tribal knowledge walks out the door when consultants leave. PII is discovered in production — a GDPR breach waiting to happen. Validation rules reject rows mid-load, blocking your weekend cutover. And every migration starts from scratch because nothing was documented.

MapFlow makes every one of these failure modes impossible. The question isn't whether you can afford MapFlow — it's whether you can afford another migration without it.

Industry Frameworks

Wider Industry Alignment

Beyond the headline standards on the product synopsis, MapFlow's controls map to these additional industry frameworks — particularly relevant for UK public-sector and enterprise engagements.

Cyber Essentials / Cyber Essentials Plus (UK Govt)

UK government-backed scheme defining a baseline of technical controls: secure configuration, boundary firewalls, access control, malware protection, and patch management.

MapFlow alignment

RBAC, MFA, session timeout, encrypted connections, and patch cadence on cloud infrastructure satisfy the five Cyber Essentials controls. Cyber Essentials Plus adds an independent vulnerability scan.

DAMA-DMBOK (Data Management Body of Knowledge)

The international standard reference for data management, covering data governance, data quality, metadata, and migration as distinct disciplines.

MapFlow alignment

Schema Catalogue, versioned mappings, audit trail, risk log, and AI profiling map to DMBOK's Data Governance, Data Quality, and Metadata Management knowledge areas.

DPA 2018 (UK Data Protection Act)

The UK's primary data protection legislation, supplementing GDPR with UK-specific provisions including intelligence services exemptions.

MapFlow alignment

Design-time PII scanning, encryption at rest, audit logging, and MFA-gated key retrieval align with DPA 2018's accountability principle (Section 8, Schedule 1).

NCSC Cloud Security Principles

UK National Cyber Security Centre's 14 cloud security principles covering data security, separation, governance, and operational security.

MapFlow alignment

Multi-tenant isolation via RLS, per-project workspaces, encrypted PII fields, and audit trail map to Principles 2 (Asset protection), 5 (Separation), and 9 (Secure user management).

NIST Cybersecurity Framework (CSF)

Voluntary US framework: Identify, Protect, Detect, Respond, Recover. Widely adopted internationally as a security maturity benchmark.

MapFlow alignment

Schema profiling (Identify), MFA + RLS + encryption (Protect), audit trail (Detect), admin alerts (Respond), backup & restore (Recover).

ISO/IEC 27017 (Cloud Security) & 27018 (PII in Clouds)

ISO extensions to 27001: 27017 adds cloud-specific controls; 27018 protects PII in public clouds acting as a processor.

MapFlow alignment

Server-side-only encryption keys, MFA-gated retrieval, per-schema PII protection, and SSL connections satisfy 27018's PII-specific controls.

OWASP Application Security

Open Web Application Security Project top-10 risks and secure coding practices for web applications.

MapFlow alignment

Server-side input validation, parameterised DB queries, no client-side secrets, and authenticated backend functions address OWASP injection, broken auth, and sensitive-data categories.

CSA STAR (Cloud Security Alliance)

Industry registry for cloud provider security posture, with self-assessment and third-party audit tiers.

MapFlow alignment

MapFlow's control documentation is structured to map cleanly onto a CSA CAIQ (Consensus Assessments Initiative Questionnaire) self-assessment.

EDM Council DCAM (Data Management Capability)

Data Management Capability Assessment Model — enterprise data management maturity framework used by financial institutions and regulated industries.

MapFlow alignment

Versioned artefacts, formal sign-off, risk register, and audit trail satisfy DCAM's Data Governance and Data Quality maturity dimensions.

UK PECR (Privacy & Electronic Communications)

UK regulations on electronic marketing, cookies, and communications — complements GDPR/DPA 2018.

MapFlow alignment

No marketing tracking without consent, analytics events are operational (no PII), and no third-party advertising scripts.

Alignment, not certification. The frameworks above describe the controls MapFlow implements. Formal certification against any of them (e.g. Cyber Essentials Plus, ISO 27001, SOC 2 Type II) requires an independent audit of the hosting environment and organisational processes — MapFlow is designed to pass such an audit, but has not yet undergone one. We are transparent about this so buyers can make an informed decision rather than infer compliance from a badge.

Independent Sources & Evidence

Not Just Our Word — The Evidence

Every headline claim on this page is grounded in independent research, industry standards, or regulatory text. These are the sources behind the numbers.

1

Gartner: Data Migration Failure Rate

Gartner research consistently finds that 50%+ of data migration projects overrun or fail to meet objectives — driving demand for higher-confidence, AI-assisted mapping.

View source
2

McKinsey: The Economic Potential of Generative AI

McKinsey estimates generative AI could add $2.6–4.4 trillion annually to the global economy, with software engineering among the top value-creating use cases — directly underpinning AI-driven discovery and mapping.

View source
3

NIST SP 800-53 Rev. 5 — AC-2 / AU-6 Audit & Accountability

U.S. National Institute of Standards and Technology control families mandate change history and audit logging — MapFlow's versioned, append-only audit trail satisfies these by design.

View source
4

Ray Panko (University of Hawaii) — Spreadsheet Error Research

Decades of research by Prof. Raymond Panko show ~88% of real-world spreadsheets contain errors, and that manual cell-entry is a leading root cause — the case for machine-generated, validated SQL.

View source
5

IBM Systems Sciences Institute — Cost of Defects by Phase

The classic IBM SSI study quantifies that a defect found in production costs ~10x or more to fix than the same defect caught at design time — the foundation of MapFlow's design-time validation strategy.

View source
6

GDPR Article 25 — Data Protection by Design

EU regulation mandates that data protection is built into processing systems by design and by default — MapFlow's AI PII scanning at design time maps directly to this legal requirement.

View source
7

NIST SP 800-63B — Authenticator Assurance Level 2

NIST digital identity guidelines define AAL2 for higher-assurance authentication; MapFlow's TOTP-based MFA meets this level.

View source
8

IBM Security — Cost of a Data Breach Report

IBM's annual report finds the average data breach costs organisations ~$4.45M, with PII exposure among the costliest record types — the financial case for design-time PII detection.

View source
9

Microsoft Access 2016+ Specifications

Microsoft's own documentation confirms the ~2 GB database file size limit for Access (.accdb) — a structural ceiling that cannot be raised without migrating off the Access engine.

View source
10

SHRM — Cost of Employee Turnover / Knowledge Loss

Society for Human Resource Management research places the cost of replacing a skilled employee at 50–200% of annual salary — the institutional-knowledge risk MapFlow's documented, versioned mappings eliminate.

View source
11

Bloor Research — Data Migration Failure Studies

Bloor Research found that the vast majority of data migration projects overrun their budgets and schedules, with inadequate data quality assessment cited as a primary cause.

View source
12

ISO/IEC 27001:2022 — Information Security Management

International standard for information security management systems; MapFlow's RBAC, MFA, session timeout, audit logging, and backup controls map to its control families.

View source
13

Salesforce — File & Data Storage Limits & Pricing

Salesforce publishes per-org file and data storage allocations; additional file storage is billed at premium per-GB monthly rates (~£9.90/GB/month), making uncompressed document uploads a persistent recurring cost — the 'Salesforce Storage Tax'. With ~85% of council documents being high-fidelity case files, this tax compounds across terabytes for the lifetime of the system.

View source

On independent auditing: None of the three tools compared on this page — MapFlow, the Arcus spreadsheet, or MS Access Datachecker — have been subjected to a formal third-party audit of their migration outcomes. The competitive claims here are industry-level findings, drawn from each tool's own published specifications and the broader research cited above.

Where a limitation is attributed to a legacy tool (e.g. the ~2 GB Access ceiling, absence of MFA, single-user desktop execution), it reflects that tool's documented technical specification or recognised industry knowledge — not a vendor-commissioned benchmark. MapFlow's own performance figures (95%+ mapping confidence, throughput) are based on observed outcomes across customer migrations and are presented alongside the third-party research so buyers can calibrate every claim against independent evidence.

Ready to migrate smarter?

The core mapping, schema, AI, governance, and overnight pipeline features are fully operational. Jump in and start documenting your migration today — or explore the synopsis and comparison pages first.