💸
Legacy middleware costs compound
BizTalk and SAP PI carry rising licence, infrastructure, and specialist labour costs. The longer the platform stays, the smaller the talent pool that can keep it running safely.
Enterprise migration guide
Migrating to a modern integration platform? Here's the playbook.
Used by Nurminen Logistics, Fazer, and Prima Pet Premium. A practical framework for moving off legacy middleware without disrupting the business that depends on it.
43%
Operational efficiency gain
3–6mo
Typical time to ROI
75%
Lower TCO vs legacy middleware
The cost of standing still
Why most enterprises move off legacy middleware now
The trigger is rarely a single failure. It's the slow accumulation of cost, fragility, and risk that makes the current platform a tax on every other strategic project.
🪢
The integration estate gets brittle
Years of point-to-point connections, undocumented orchestrations, and one-off fixes turn the platform into a fragile core that blocks every new business initiative.
⏳
Vendor end-of-life dates are real
SAP PI/PO mainstream support ends in 2027. BizTalk 2020 is the last release. Waiting moves the migration from a planned programme to an emergency one.
Real migrations
From spaghetti to clarity
Three organisations. Three different strategies. Each maps to a chapter of the playbook — read how the theory lands in production.
Featured · Big bang migration
Nurminen Logistics
From Microsoft BizTalk plus two other platforms
“Frends is now the reliable core of all our data — the digital integration hub our business actually runs on.”
CIO, Nurminen Logistics
Platforms
3→
1
Efficiency
Baseline→
+43%
Reused assets
0→
BizTalk C# + maps
Starting point
Three integration platforms running in parallel, no in-house integration team, escalating maintenance bills.
What they did
Frends' .NET-native architecture let the team reuse existing BizTalk assets — C# code, transformations, orchestration logic — without rewriting from scratch. Three platforms became one. The CIO describes Frends as the reliable core of all data and the digital integration hub for the business.
Lesson for your migration
Commit fully. Half-migrations leave you paying twice — once for the legacy, once for the new.
Fazer
From SAP PI plus EDI tools and finance systems
Phased by domain3→1
Platforms consolidated
“Each wave produced reusable components — the next wave was always faster than the last.”
Integration Lead, Fazer
Starting point
SAP PI end-of-life across eight countries with autonomous business units pulling in different directions.
What they'd flag for you
Sequence waves so the first one builds the templates the rest will reuse — EDI first paid off three times over.
Prima Pet Premium
Greenfield iPaaS adoption
Phased by criticalitySelf-sufficient
Team builds and maintains independently
“By wave three we were operating without consultants — that was the goal from day one.”
Head of IT, Prima Pet Premium
Starting point
Small IT team, zero iPaaS experience, ERP that could not afford a single hour of disruption.
What they'd flag for you
Pilot something low-stakes first. The team learning the platform is more valuable than any quick win.
Interactive tool
Find your migration strategy
Five questions. A personalised recommendation based on your portfolio size, team experience, and risk appetite — with a matching case study.
Question 1 of 5
How many integrations does your organisation currently run?
Select an option
Big bang migration
Single cutover event. Best for smaller portfolios with strong testing capability. Fast, but all-or-nothing.
Phased by domain
Move by business domain — CRM first, then ERP, then Finance. Cleaner risk control when domains have clear boundaries.
Phased by criticality
Start simple, build confidence. Low-risk integrations first — the team learns the platform before tackling mission-critical systems.
Interactive tool
Integration criticality matrix
Plot your integrations by business impact and technical complexity. The quadrant they land in determines their migration priority and wave sequence.
Each quadrant is one phase of your migration — quick wins first, deferred work last. Hover the i on a quadrant for guidance.
↑ Business impact
Quick winsWave 1
High impact · Low complexity
Customer data sync
Email notifications
Critical pathWave 2–3
High impact · High complexity
ERP order processing
Financial reporting
Easy pickingsWave 3–4
Low impact · Low complexity
Legacy file transfers
Defer or simplifyWave 4+
Low impact · High complexity
Custom connector (outdated)
Technical complexity →
Suggested migration sequence
Start with your 2 quick wins in Wave 1 — these deliver immediate value and build team confidence. Then plan extended parallel running for your 2 critical path integrations in Waves 2–3 with detailed rollback procedures. Evaluate the 1 lower-impact item for retirement before committing migration effort.
Before you cut over
Get your data ready
The part of migration teams routinely underestimate. Most slipped timelines trace back to data work that wasn't done early enough — not to platform problems.
✓Do this first
🗂️
Audit your integration inventory
List every flow, including the cron jobs nobody owns. You cannot migrate what you cannot see.
👤
Map data ownership
Every entity needs a named business owner who signs off on its shape, not just a system that stores it.
💎
Define golden records
Decide which system is the source of truth for customer, product, and order before migration day, not during cutover.
🏷️
Version your schemas
Tag every contract. Migration is the moment breaking changes surface — versioning lets you fix them without a fire drill.
🧪
Dry-run with production-shaped data
Sanitised copies of real volumes catch issues synthetic test data never will. Allocate time for at least two full dry runs.
📈
Baseline current SLAs
Measure latency, throughput, and error rates today. Without a baseline you cannot prove the new platform is better.
✕Avoid these traps
📦
Lift-and-shift without cleanup
Migrating bad integrations one-for-one just moves the mess. Use the migration to retire what no business owner can defend.
💻
Treating migration as a code project
It's a business change project with code in it. Without business owners in the room, scope drift is guaranteed.
↩️
No rollback plan
Every cutover needs a documented rollback that has been rehearsed, not just written down. Hope is not a recovery strategy.
⏸️
Skipping the parallel-run window
Run old and new in parallel for at least one full business cycle. Cutover with no comparison data is a guess.
🌙
Hidden batch jobs
Nightly scripts on a server under someone's desk. They surface at 02:00 on go-live. Hunt them down weeks in advance.
🤐
Undocumented tribal integrations
If the only person who understands a flow is on holiday during cutover, that flow will break. Document or rebuild before they leave.
“Data debt is migration debt. Whatever you didn't clean up before cutover, you will clean up during — at three times the cost and with the business watching.”
From the Frends migration strategy whitepaper
Interactive tool
Migration readiness checklist
Work through the five readiness dimensions before kick-off. Your score updates as you check items off, and the share link lets your team complete it together.
0%
Overall readiness score
Complete items to assess readiness
0 of 25 items complete
After migration
Migration is the starting line
Organisations that treat migration as the finish line watch their new platform drift into the same chaos they just escaped. Here's what the best ones do next.
⚖️
Governance without bureaucracy
Standards that enable speed. A living integration handbook — naming conventions, approved patterns, security requirements — that developers actually use because it makes their work easier.
🧩
Enable, don't centralise
One central team doing all the work creates backlogs. A Center for Enablement publishes reusable templates and trains business units to build independently, scaling capacity without scaling headcount.
🤖
Adopt AI strategically
Start with high-volume, low-risk use cases — invoice exception handling, ticket routing. Keep transparency and auditability. Pilot, learn, refine, then scale. Not everything needs AI, and not all at once.
📊
Monitor intelligently
Define normal baselines. Configure alerts that distinguish signal from noise. Build dashboards that show integration health in business terms, not just uptime percentages.
🔄
Control technical debt
Schedule regular refactoring sprints before debt accumulates. Update deprecated connectors, consolidate redundant integrations, forecast capacity before it becomes an incident.
🤝
Treat your vendor as a partner
Quarterly roadmap reviews, beta program access, active feedback to shape platform direction. The organisations that get the most value treat their iPaaS vendor as a strategic relationship, not a software purchase.
Get moving
Generate your migration plan
Three quick inputs, a phased wave plan tuned to your team size, and an export you can drop straight into Google Calendar, Outlook, Jira, Asana, or Trello.
DD/MM/YYYY · 15/08/2026
Strategy
5
SoloFull squad
Preview
4 waves · 164 working days
15/08/2026 → 26/01/2027 · scaled 1.15× for team size
This is a preview using your current inputs. Hit “Generate plan” to lock it in and enable the calendar and CSV exports.
Ready to map out your migration?
Talk to a Frends migration specialist. We'll review your integration landscape, recommend a strategy, and scope what a migration to Frends actually looks like for your organisation.