The Intermediary –- May 2026 - Flipbook - Page 97
T E C H N O L O GY
Opinion
societies are
decisions. On a genuinely composable
platform, that’s a relatively simple
configuration change.
To work out if a core is truly
composable, there’s a simple twoquestion test: can your product team
use best practice modules to create
new products, without developer
time? And can you make a product
change without triggering a cascade of
regression testing across the system?
If the answer to either of these is no,
you don’t have composability.
2. Zero-downtime upgrades
A genuine fourth-generation core
must support zero-downtime upgrades
and releases, and continuous release
cycles should come as standard. For
many platforms on the market, they
don’t. Upgrades may be technically
possible on older or pseudo-modern
systems, but each modification
increases regression risk.
Institutions delay version adoption,
not for lack of ambition, but because
structural revalidation is too costly. In
that event, legacy doesn’t disappear, it
just gets containerised.
What this should look like
in practice is: monthly feature
releases, a single platform version
across all clients, and continuous
daily deployments without
scheduled downtime.
For building society members, that
translates to no service disruption,
access to the latest products as they
arrive, and security improvements
applied continuously rather than in
periodic, disruptive cycles.
3. AI-ready data flows
We all know how artificial intelligence
(AI) will increasingly reshape banking
from the inside, with the likes of
predictive servicing, dynamic pricing,
real-time credit decisioning, and
proactive risk intervention.
All of these capabilities depend
entirely on the data that flows from
the core banking platform you use.
AI needs clean, real-time data flows
and clearly exposed domain events.
Mutuals that can harness real-time
member data will be beer placed
to deliver the kind of personalised,
values-led service that differentiates
them from big banks.
If you’re thinking about your
organisation’s AI readiness, the
very first thing you need to know
is whether or not your core’s
architecture can support AI in the
first place.
A modern core
The conversation about core
modernisation in the mutual sector
tends to get stuck on infrastructure
costs: what does the platform cost
to run? What does it cost to change?
When does the investment pay back?
It’s an understandable frame, but it
misses most of what’s actually at stake.
Building societies that have moved
to genuinely modern infrastructure
describe a qualitatively different
experience of running their
organisation. Instead of products
taking months to launch, you can do
it in days, and deliver rate changes
without member-facing disruption.
And you can react in near real-time to
regulatory requirements.
Strategically speaking, a modern
core gives a building society the
ability to move when the market
moves, rather than just when the
platform allows.
That means being able to explore
embedded finance and reach members
through third-party channels, rather
than waiting on a vendor’s roadmap. It
means having the data infrastructure
to build genuinely personalised
member experiences, rather than
retrofiing intelligence onto a system
that was never designed for it. And it
means not fighting the platform every
time a new product idea or regulatory
requirement arrives.
For mutuals, with their long-term
view and member-first obligations,
that kind of structural freedom is both
commercially valuable and central to
their mission in the first place.
The simple test
The question I encourage every
mutual to ask is this: show me a zerodowntime upgrade, in production,
at scale. Show me a product
launched through configuration, not
redevelopment. Show me how your
event architecture makes real-time
intelligence possible, in practice.
Most platforms can demonstrate
functionality fine, but you need
to know if they can demonstrate
these three things under realistic
operating conditions.
A well-designed proof of concept
is how you find out, testing the
engineering constraints that will
define your freedom to operate five,
10, 15 years from now.
Anyone who knows me knows I love
my analogies and frameworks. But
‘the trinity’ isn’t something I invented
to fill a slide. It’s the minimum I’d
want verified before investing in
a technology that will define how
an institution operates for the
next decade.
Get a core platform that can
demonstrate all three in production.
If they can’t or won’t, you have your
answer, and you’ve avoided building
your members’ future on a Spinal Tap
Stonehenge. ●
May 2026 | The Intermediary
95
