Specialist service · software modernization services

Legacy-to-SaaS modernisation

We usually migrate in stages using the strangler-fig pattern: new modules are built as SaaS services alongside the old system, traffic moves module by module, and the legacy code is retired last.

Product Engineering practice: abstract red and ink artwork from the Red Balloon reel

Legacy software that is installed separately at every client is costly to support and hard to sell. Red Balloon's software modernization services help product companies move ageing desktop or on-premise systems to a modern, multi-tenant SaaS platform without breaking the customers who rely on them every day. We start by mapping what the old code really does, writing tests that pin down current behaviour and planning data migration client by client. Business rules buried in old code are preserved deliberately, not rediscovered after go-live. Once clients run on one platform, you ship one release to everyone instead of patching each installation. This work suits software vendors in distribution, banking, education and manufacturing.

What drives the cost and timeline of modernisation

The size of the codebase matters less than how far the installations have drifted apart. A product with one version at every client is a far smaller job than one with years of client-specific changes, custom reports and undocumented database columns. The other big drivers are data volume and quality, the number of business rules buried in stored procedures, integrations with banks, tax systems or hardware, and how many clients need a parallel run before cutover.

Timeline is also shaped by your clients, not just your code. Each migration wave needs pilot clients willing to test, staff to train users, and cutover dates that avoid their financial year-end or peak season. Our software modernization services start with an assessment that maps all of this, then give you a phased plan and a quote per wave rather than a single number for the whole programme.

Repackaging the product once it runs as SaaS

Moving to SaaS changes how you sell as much as how you build. Many Indian software businesses still earn through one-time licences plus an annual maintenance contract, with upgrades priced separately. A multi-tenant platform makes subscriptions, usage tiers and module-based plans possible, but existing clients need a clear reason to move and a fair path from what they pay today.

Our Launch & Go-to-Market practice works on this alongside the engineering. That covers plan structure, what is included at each level, how existing contracts transition, and how sales and support explain the change. Product decisions such as feature flags per plan and billing integration are built in from the first module, so pricing changes do not need code changes later.

Keeping your existing developers at the centre

The people who maintain the legacy system know where its rules and quirks live, and losing them mid-programme is one of the biggest risks in any modernisation. We plan the work so they stay central. Long-serving developers help write the characterisation tests, review how old behaviour is reproduced, and explain the edge cases that no document records.

At the same time they learn the new stack by building with it. Pairing on the first modules, code reviews and short internal sessions on the new architecture mean that by the final wave your own team is shipping features on the SaaS platform. The goal is a team that owns the new product, not one that feels replaced by it.

Preparing for the security questions enterprise buyers ask

Once your product is cloud-hosted, larger clients will send security questionnaires before renewing or signing. They will ask where data is stored, how tenants are isolated, who can access production, how backups are tested, and how incidents are reported. Answers that were easy for an on-premise install become your responsibility as the provider.

We build the answers into the platform: tenant isolation tested by automated checks, role-based access with audit logs, single sign-on for enterprise clients, encrypted backups with restore drills, and infrastructure defined as code. If you plan to pursue ISO/IEC 27001 or a SOC 2 report, these controls give the auditor evidence rather than promises. As a provider of software modernization services in India, we document each control so your team can answer questionnaires without calling engineering every time.

Where this works

Real-world situations we're built for

Illustrative · not a client
Pharma distribution software

Delhi vendor's Visual Basic desktop app is installed separately at 300 distributor sites

What we'd do
Strangler migration to a multi-tenant web SaaS, starting with reporting and order entry
What moves
Support visits down, versions in the field down; tracked as versions in the field before and after, over a fixed period
Co-operative banking software

Maharashtra provider maintains a separate on-premise database for each co-operative bank client

What we'd do
Staged move to a cloud platform hosted in Indian regions with tenant isolation, module by module
What moves
Release cycle time down, client upgrade lag down
How a project runs

4 phases, each with a deliverable

What happens, how long it takes, what you get and what we need from you.

Assess

2–4 weeks

Code, data and infrastructure audit, dependency map, option scoring

You get: Modernisation roadmap

We need: Source code, database access, client list

Foundation

4–8 weeks

Target platform, tenancy, CI/CD, integration bridge to the legacy system

You get: Platform running the first module

We need: Pilot clients

Migrate in waves

3–9 months

Module rebuilds, data migration, parallel runs

You get: Modules live wave by wave

We need: Client communication and acceptance testing

Retire legacy

1–2 months

Decommissioning, archiving, final data checks

You get: Legacy system shut down

We need: Client sign-offs

A typical engagement

From first call to result

Illustrative · not a client

Gurugram ERP vendor, 120 client installs of a .NET Framework and SQL Server desktop app, 15 developers

Weeks 1–3AssessmentCode and database audit, client version census, module dependency map; check first which stored procedures hold GST and pricing logicMigration roadmap by module
Weeks 3–10Platform foundationMulti-tenant web platform, auth, tenant configuration, data sync bridge from desktop SQL ServerReports module live for 5 pilot clients
Weeks 10–30Module wavesOrder entry, inventory and billing rebuilt one by one; desktop app reads the new APIs meanwhileEach wave live for pilots, then all clients
Weeks 20–34Client data migrationPer-client data moves with row counts and checksums per table, a parallel-run weekend, cutover only at zero unresolved mismatchesClients cut over in batches
Weeks 34–40DecommissionDesktop installs retired, old databases archived under retention rulesOne version for every client
Where it starts9 versions live across clients, 40% of support time spent on upgrades, new features take 6 months to reach everyone
Where it endsAll clients on one multi-tenant version. Weekly watch: clients migrated, data validation errors, support tickets per client.
What we measure

The numbers we report on

Every metric has a formula, a target, a source and a reporting rhythm.

MetricHow it's calculatedTargetSourceCadence
Migration progressClients or modules on the new platform ÷ totalTrack against the planned roadmap per waveProject trackerWeekly
Data validation errorsRecord mismatches after migration ÷ records migratedZero unresolved before each cutoverValidation scripts or dbt testsEach migration batch
Deployment frequency (DORA)Production releases per month on the new platformAt least weekly releases to all tenants (rule of thumb); DORA reports elite teams deploying on demandCI/CD logsMonthly
Change failure rate (DORA)Releases causing an incident ÷ total releasesSet from baseline, then aim downwardIncident log and CI/CD recordsMonthly
Automated by default

What runs itself after launch

The manual work we take out of this service, and the tools we use.

When: Each client cutoverData migration with checksAWS DMS, Airbyte or custom ETL with row counts and checksumsRemoves: Missing or corrupted records discovered later
When: Before each module migrationCharacterisation testsAutomated tests capturing current legacy outputs, such as invoice totals for 500 real orders, before a module is rewrittenRemoves: Silent changes in calculations such as GST or pricing
When: Merge to mainSingle release to all tenantsCI/CD with feature flags per tenantRemoves: Per-client upgrade visits
Rules we work to

Standards and regulations

ISO and IECISO/IEC 27001:2022

A modernised SaaS will face buyer security reviews; build controls in rather than retrofit

MeitYDPDP Act 2023 and DPDP Rules 2025

Moving client data into a shared cloud needs processor contracts, safeguards and retention rules; main duties apply from 13 May 2027

CERT-In (MeitY)CERT-In Directions of April 2022

Cloud service providers must report specified incidents within 6 hours and retain logs for 180 days

What's changing

Platform and policy updates that affect this work

We track: Microsoft Lifecycle announcements, endoflife.date.

CentOS ProjectCentOS Linux 7 reached end of life

Legacy servers on CentOS 7 no longer receive security updates; a clear migration trigger

MicrosoftWindows 10 reached end of support

Desktop software running on Windows 10 machines carries rising security risk

Before you hire anyone

Questions to ask any agency, including us

Which module would you migrate first, and why?

Why ask

The first wave sets risk and momentum for the whole programme

A good answer

A high-value, loosely coupled module, justified from the dependency map

How will you find business rules nobody documented?

Why ask

Legacy code hides rules like discounts and tax logic

A good answer

Code reading, characterisation tests and sessions with long-serving users

How will our existing clients be informed and supported?

Why ask

Client churn during migration is a real risk

A good answer

A communication plan, pilot programme, training and support during cutover

FAQ

Questions buyers ask us

Straight answers from the practice lead. Highlighted text marks Red Balloon figures we confirm before publishing.

How do you migrate data from an on-premise legacy database without downtime?

We copy historical data first, stream ongoing changes with change data capture until the new system is in sync, then cut over during a short, rehearsed window with a rollback plan.

  • AWS DMS or Debezium handle change data capture from SQL Server, Oracle or MySQL.
  • We rehearse the cutover at least once on a copy of production and reconcile row counts, totals and sampled records; rehearsals usually surface date-format and character-encoding bugs.
How do we move separate client installations onto one multi-tenant platform?

Each client's customisations are catalogued and turned into configuration, feature flags or extension points, then clients move one at a time onto a single version, starting with the least customised.

  • Customisations used by one client that no longer pay for themselves are candidates for retirement.
  • Client contracts may need updating for data location, uptime commitments and upgrade schedules.
  • Clients on very old versions may need a data transformation step before migration.
  • Compare database schemas across installations as well as the code; undocumented columns added for one client are where surprises hide.
Can you modernise software written in VB6, Delphi or old .NET?

Yes. Age of language is rarely the blocker; missing knowledge is. We read the existing code, run it against real data to capture its outputs, and write characterisation tests before anything is rebuilt. Where the original build tools are hard to run, we set up a working legacy environment first, so the old system can be compared with the new one throughout the migration.

  • Legacy code read and tested first
  • Outputs captured from real data
  • Old and new compared side by side
What happens to clients who do not want to leave the on-premise version?

There are usually three options. Set a clear end-of-support date with enough notice and help to move; offer extended support at a price that reflects its cost; or, for large clients with strict data needs, run a single-tenant deployment of the new platform in their own cloud account. We help you choose per client and write the communication, so the decision is handled early rather than at cutover.

  • Clear end-of-support date
  • Paid extended support
  • Single-tenant option for large clients
Back to the service

SaaS development →

Start with an audit

Ready for
lift?

A paid two-week Growth Audit covers your funnel, ads, site, tracking and automation gaps, and ends with a 90-day plan you can run with or without us. The fee is credited against your next engagement.

Prefer to talk? Call +91 97114 48083, or email info@redballoon.in.

A person from the team replies to every enquiry.

What do you need?
Your details are sent over a secure, encrypted connection and only used to reply to you.