Rehost
lift-and-shiftMove the workload to AWS with no code changes. Infrastructure changes; application does not.
saasups looks at
High dependency count, low change frequency, tight migration timeline, no cloud-native requirement from the business.
THE PIPELINE
From estate data to executable roadmap — four precise steps.
saasups runs a deterministic pipeline. No workshops. No interviews. No consultant opinions baked into the model.
01
Ingest the estate
saasups connects to your running systems — cloud accounts, container registries, IaC repositories, service meshes, and config stores. It reads what is actually running, not what documentation says should be running.
Supports AWS, on-prem, hybrid, and containerised workloads.
02
Map every dependency
The dependency graph engine traces every inbound and outbound service call, database connection, message queue binding, and shared-state coupling. Circular references are resolved. Undocumented connections surface.
The graph is the foundation. Every downstream score and recommendation is derived from it.
03
Score against all 7 R's
Each workload is evaluated against all seven modernization strategies — Rehost, Replatform, Refactor, Rearchitect, Rebuild, Replace, Retire. The scoring model weights dependency count, change frequency, business value, cloud fit, and compliance constraints.
Every score is ranked, not binned. You see the best strategy and the full ranked list.
04
Generate the executable roadmap
saasups sequences your scored workloads into a dependency-ordered migration pipeline, embeds FinOps cost modeling for each phase, and outputs artifacts your engineers can act on — not a deck your board has to interpret.
Roadmap artifacts include wave plans, risk flags, and before/after cost projections.
THE SEVEN STRATEGIES
One framework. Seven precise outcomes.
The 7 R's are not a checklist — they are a decision surface. saasups scores every workload against all seven before committing to a recommendation.
Replatform
lift-and-reshapeMove to AWS with targeted runtime or managed-service substitutions — no core logic rewrite.
saasups looks at
Moderate dependency count, managed-service substitution available (e.g. self-managed DB → RDS), low refactor risk, positive TCO delta.
Refactor
re-architect incrementallyRestructure existing code to improve cloud fit without a full rebuild — typically breaking a monolith into services.
saasups looks at
High change frequency, bounded context identifiable in the codebase, moderate coupling score, engineering team capacity confirmed.
Rearchitect
redesign for cloud-nativeRedesign the application's core architecture to exploit cloud-native patterns — event-driven, serverless, or microservice-native.
saasups looks at
Low dependency coupling, high business value, cloud-native runtime available, application owned by team with cloud competency.
Rebuild
rewrite from scratchDiscard the existing implementation and rewrite the capability in a cloud-native stack.
saasups looks at
Technical debt density exceeds refactor threshold, no viable incremental path, capability is strategically critical, team capacity and timeline allow.
Replace
swap for SaaSRetire the custom application and adopt a SaaS equivalent that covers the capability without the operational burden.
saasups looks at
Commodity capability, low differentiation score, SaaS equivalent with required compliance posture exists, TCO favours SaaS over migration cost.
Retire
decommissionDecommission the workload. No migration, no replacement — the capability is not needed.
saasups looks at
Zero or near-zero active users, no inbound dependencies from live systems, business owner confirms capability is obsolete.
SCORING MODEL
Every recommendation is auditable. None are opinions.
saasups derives scores from five measurable input dimensions. Each dimension is weighted by the scoring model, and the weights are visible — not hidden inside a black box.
Each scored workload carries a full evidence trail: which data sources contributed, what weight each dimension received, and why the top-ranked strategy scored above the alternatives. Every recommendation is traceable back to your estate data.
Estate inputs
Weighted scoring
Ranked recommendation + evidence
Dependency topology
Inbound and outbound coupling count, transitive depth, circular reference count, and shared-state surface area — derived from the live dependency graph.
Change history
Commit frequency, deployment cadence, incident rate, and rollback frequency — pulled from your VCS, CI/CD, and incident log.
Traffic & usage
Request volume, peak concurrency, SLA tier, and active-user count — from APM, load balancer logs, or cloud-native metrics.
Cloud-fit assessment
Runtime compatibility with AWS managed services, container readiness, IaC coverage, and security posture against AWS Well-Architected criteria.
Compliance constraints
Data residency requirements, regulatory classification, encryption-at-rest status, and audit log completeness — mapped to the target landing zone.
ILLUSTRATIVE OUTPUT
A sample scored estate — what saasups produces.
Illustrative only. Application names, scores, and rationale are constructed to demonstrate output structure — not derived from any real estate.
ApplicationTypeRecommended RScoreConfidenceRationale
billing-coreJava monolithRefactor87/100HighHigh change frequency, bounded billing context identifiable, coupling score within refactor threshold.
auth-gatewayNode.js serviceReplatform82/100HighLow coupling, session store replaceable with ElastiCache, positive TCO delta confirmed.
report-schedulerPython batch jobRetire94/100HighZero active users confirmed, no inbound live dependencies, business owner decommission approval on file.
customer-portal.NET Web FormsReplace76/100MediumCommodity capability, low differentiation, SaaS equivalent (Salesforce Experience Cloud) passes compliance check.
inventory-engineC++ legacy serviceRebuild71/100MediumTechnical debt density exceeds refactor threshold, no viable incremental path, strategic capability confirmed.
event-ingestorKafka consumerRearchitect89/100HighLow coupling, event-driven pattern cloud-native ready, team has Kinesis/Lambda competency confirmed.
FREQUENTLY ASKED
Questions engineering leaders ask before the first assessment.
The "6 R's" formulation omits Rebuild — treating a ground-up rewrite as a subset of Refactor. That conflation obscures the most expensive decision in modernization. saasups scores Rebuild separately because the signals that trigger it (debt density above the refactor threshold, no viable incremental path, strategic criticality) are distinct from Refactor signals. Collapsing the two produces incorrect recommendations for workloads where a rewrite is genuinely the right call.
Yes. saasups produces a ranked recommendation with full evidence — it does not issue mandates. Engineering teams can override any recommendation by adjusting constraint inputs (timeline, team capacity, compliance requirements) and re-running the scorer. Every override is logged against the original recommendation so your audit trail stays intact. The system is designed to support human judgment, not replace it.
Each workload is scored across five dimensions: dependency topology, change history, traffic and usage, cloud-fit assessment, and compliance constraints. Each dimension is weighted by the scoring model. The model outputs a ranked list of all seven strategies — not a single verdict — with the top-ranked strategy carrying the highest composite score. The full weight breakdown is visible in the assessment output. saasups does not use a hidden scoring formula.
Initial estate ingestion and dependency graph construction typically completes within 48–72 hours for a workload of 50–200 applications, depending on connector availability and estate complexity. Scoring runs continuously as estate data updates. Most engineering leaders see a complete scored estate within the first two weeks of assessment — versus the six-to-twelve weeks a consulting assessment requires.