Your technology strategy is a switching strategy
Nine months. That's how long a mid-market technology project typically takes. About the same time it takes to make a new human.
Your technology strategy is a switching strategy
Nine months. That's how long a mid-market technology project typically takes from the first scoping document to something running in production. Requirements gathering, vendor evaluation, procurement, implementation, testing, training, go-live. Nine months. About the same time it takes to make a new human.
In those nine months, three of your competitors switched CRM. One of them rebuilt their entire client onboarding flow in six weeks using tools that didn't exist when you started your project. The market moved. You're still in UAT.
The strategic question isn't "which system should we use?" It's "how fast can we stop using this one and start using that one?"
The old model is a trap
Traditional technology strategy treats systems as long-term commitments. You pick a platform, you implement it, you build around it, you live with it for five to ten years. The switching cost is so high that you'd rather patch and extend than replace. So you end up with a stack that's three years behind the market and six months away from a crisis.
That model made sense when software was installed on servers you owned. It doesn't make sense when every serious business tool is a SaaS product with monthly billing and an API.
Good architecture lets you switch quickly
The difference between a business that can move and one that can't isn't budget or team size. It's architecture.
Good architecture means your data isn't trapped. It means your integrations connect through documented APIs, not spreadsheet imports and manual re-keying. It means your processes are recorded somewhere other than the head of the person who set them up four years ago.
When you need to replace a component — and you will — good architecture makes that a project measured in days, not quarters. We migrated a CRM last month. The system swap took a day. The data work took nine. Total elapsed: two weeks. The previous system had been "about to be replaced" for three years.
Great architecture is a weapon
Good architecture removes friction. Great architecture creates advantage.
If you can evaluate, test, and switch to a better tool faster than your competitors, you're not just keeping up — you're compounding. Every switch is a chance to pick the best current option, configured for where your business is going right now, not where it was when the original project kicked off.
Partner selection matters here too. The vendor landscape shifts constantly. A tool that was best-in-class eighteen months ago might be stagnant today while a competitor shipped the features you actually need. If your architecture lets you switch without a nine-month programme, you can pick partners based on current capability rather than sunk cost.
That's not operational efficiency. That's strategic advantage. The business that can switch fastest gets to choose the best tools at any given moment. Everyone else is locked into yesterday's decision.
What this actually requires
None of this is theoretical. It comes down to specific choices:
Your data must be portable. If you can't export everything in a usable format, you don't own it — you're renting access to it. Check this before you sign, not when you want to leave.
Your integrations must be replaceable. If swapping one tool means rewiring six others by hand, your integration layer is the bottleneck. API-first platforms connected through an automation layer — Make, Zapier, n8n — give you the ability to redirect data flows without rebuilding from scratch.
Your processes must be documented. Not in a 200-page playbook. In the configuration itself — named fields, clear workflows, config-as-code where the platform supports it. When the next person picks this up, they should be able to understand what it does and why.
The question to ask
Next time someone proposes a technology project with a nine-month timeline, ask this: what happens if the tool we pick today isn't the right one in twelve months?
If the answer involves another nine-month programme, your strategy isn't a strategy. It's a bet.
And you're betting that the market will stand still while you catch up.
It won't.
We wrote a case study showing what this looks like in practice.
We help mid-sized businesses build technology stacks designed for switching speed — not vendor lock-in. If your current setup takes quarters to change, book a coffee with our founder.
Robin Carswell
More on
When you need CTO-level thinking but can't justify the hire
Most mid-sized businesses don't need a full-time CTO. They need experienced technology thinking, available consistently, at a scale that fits where they are now.
What should you actually do about AI?
Most founders told to "get an AI strategy" haven't identified which problem they want it to solve. That's usually the better place to start.
Where does your data actually live?
When you sign up to a SaaS tool, your data goes somewhere. Most UK businesses haven't decided where — and clients in legal, healthcare, and financial services will ask.