The verdict in three sentences
Refactor when the debt is localized and the architecture still sound; rewrite when the technology is obsolete, maintenance ruinous and business risk high. Refactoring costs 20 to 50% of a rewrite but leaves the underlying debt; a rewrite costs 50,000 to 250,000 EUR but starts on fresh foundations. In 2026, the winning middle path is the strangler pattern: replace progressively, module by module, with no big bang.
The decision grid
Before deciding, rate your application on these axes. The more warning signals stack up, the more a rewrite (or strangler) is justified over a simple refactor.
| Criterion | Leans toward REFACTOR | Leans toward REWRITE |
|---|---|---|
| Stack age | < 6 years, maintained | Obsolete, unsupported |
| Technical debt | Localized | Diffuse, systemic |
| Maintenance cost | Stable | Rising, > 25% of build/yr |
| Available skills | Easy to hire | Rare, expensive |
| Security risk | Under control | Unpatchable flaws |
| Ability to evolve | Still possible | Blocked, each change breaks |
Compared costs and the strangler approach
The decision is first economic. Here is a 2026 order of magnitude for a mid-sized business application whose full rewrite is estimated at 150,000 EUR.
| Approach | 2026 cost (EUR) | Timeline | Residual debt | Risk |
|---|---|---|---|---|
| Targeted refactor | 30,000 – 75,000 | 2 – 4 months | Partial | Low |
| Strangler pattern | 90,000 – 160,000 | 6 – 12 months | None at end | Moderate |
| Big bang rewrite | 150,000 – 250,000 | 6 – 12 months | None | High |
| Status quo (do nothing) | 40,000+/yr maint. | — | Growing | Growing |
The strangler pattern combines the best of both: you wrap the legacy, redirect traffic module by module to the new system, and the old one fades out progressively. Big-bang risk disappears, value arrives continuously.
Mini case study
Sonia, CIO of an insurer in Abidjan, runs a legacy app whose maintenance costs 45,000 EUR/yr and where each change takes triple the normal time. A full rewrite is quoted at 180,000 EUR as a big bang, judged too risky. By opting for a strangler pattern over 10 months at 150,000 EUR, she first migrates the underwriting module (40% of traffic), then claims management. As soon as phase 1 ends, annual maintenance drops from 45,000 to 26,000 EUR, saving 19,000 EUR/yr while migration continues without service interruption.
FAQ
Need a professional website?
Kolonell builds websites that attract clients, optimized for the Sénégalese market. Free quote in 2 minutes.
When is a simple refactor enough?
When debt is localized in a few modules and the stack remains supported. You clean those zones without rebuilding everything, for 20 to 50% of a rewrite's cost.
Why is a big-bang rewrite risky?
Because it replaces everything at once, without a net: regressions, delays and service outage. Historically, a significant share of big-bang rewrites fail or overrun.
What is the strangler pattern?
A strategy where you surround the legacy with a new system and migrate traffic module by module. The old code fades out progressively, without a break.
Is doing nothing an option?
Rarely viable: the status quo often costs 40,000 EUR+/yr in rising maintenance, plus security risk and the inability to evolve. Inaction has a real price.
How do we decide objectively?
Cross annual maintenance cost, stack obsolescence and business risk. If maintenance exceeds 25% of the build cost per year, modernization becomes a priority.
Let's scope your project. Describe your legacy app (stack, maintenance cost, blockers): we propose a costed decision grid and a phased strangler plan. Detailed quote within 48 h. WhatsApp +221 77 596 93 33.
Mohamed Bah
Fondateur, Kolonell
Passionate about digital and entrepreneurship in Africa, Mohamed has been helping Sénégalese businesses with their digital transformation since 2020. Founder of Kolonell, he believes every SME deserves a professional and accessible online présence.
