Legacy modernization·4 min read
The alternative to a big-bang rewrite has had a name for twenty years
Fowler named the pattern after a plant that grows around its host and replaces it in place. Most failed replatforming programmes are ones that ignored it.
Source
Strangler Fig Application
Martin Fowler · martinfowler.com · 2004, rewritten 2024
Read the originalAslanWay is not the author of this work. The summary below describes the source; the commentary that follows is ours and is not endorsed by the author.
What it argues
Martin Fowler took the metaphor from strangler figs in Queensland: the fig seeds in the branches of an existing tree, grows down around it, and eventually stands on its own where the host used to be. Nothing is ever demolished and rebuilt from nothing.
Applied to software, you put an interception layer in front of the legacy system, then move capabilities behind it one at a time. Both systems run in parallel; traffic shifts gradually; the old system shrinks until it can be switched off.
Fowler rewrote the page in 2024 to emphasise the mechanics rather than the metaphor — the interception point, how state is handled across the seam, and how you decide the order in which capabilities move.
The pattern's real argument is about risk. A rewrite defers all value and all risk to a single cutover date. A strangler fig delivers value continuously and keeps every step reversible.
What we do with it
- The hardest part is rarely the routing layer — it is shared state. Decide early whether the seam sits in front of the data or behind it, because retrofitting that decision is what sinks these programmes.
- Sequence by rollback cost, not by business excitement. Read-heavy paths like catalogue and search go first because reverting them is nearly free. Payments and ledger go last.
- The pattern also solves a commercial problem: it removes the feature freeze. That is usually what makes the business case work, because the freeze often costs more than the migration.