Strategic planning for systems, not software
Software plans go stale the moment a vendor changes its pricing. Systems plans survive it. Here is how to plan for the thing that lasts.
By Mike Beasley, Founder, Mike Beasley & Associates
What this says
- Plan the flow of information, not the list of tools. Tools are replaceable, flows are not.
- Sequence so every step stands alone; a step that only pays off after the next one is a bet, not a phase.
- Decide what you will never automate before you decide what you will.
Most technology plans are really software plans wearing a strategy costume. They name products, versions and vendors, and they go stale the moment one of those vendors changes its pricing, gets acquired, or discontinues the tier you built on.
A systems plan is a different document. It describes how information moves through your organisation, which handoffs are automatic and which are somebody’s job, and what you intend to do about that. It survives a vendor change because it never depended on the vendor in the first place.
Plan the flow, not the tools
The instinct to plan in products is understandable. Products have prices, so they fit in a budget, and they have names, so they fit in a sentence. Flows have neither, which is why they get skipped, and why the plan fails.
Try it the other way. For each thing your organisation does repeatedly, write the journey of the information rather than the name of the software it currently lives in. An enquiry arrives, becomes a contact, gets a proposal, becomes a project, generates an invoice, gets paid, gets closed. That sequence is stable. Which product sits at each step is not.
Now you have something durable. When a tool changes, you re-point one step. When a tool’s price doubles, you can swap it without rewriting the plan, because the plan was never about the tool.
Sequence so every step stands alone
This is the discipline that separates a plan from a wish. Every phase should deliver something useful in the week it ships, independent of whether the next phase ever happens.
It is tempting to design a beautiful end state and then slice it into phases. Do not. A phase whose value only arrives once three more phases are done is not a phase, it is a bet. If the budget or the appetite stops halfway, you are left with the cost and none of the benefit.
- Good phase: the enquiry form now writes straight into the CRM, so nobody re-types anything. Useful on day one.
- Bad phase: the enquiry form is rebuilt to a new standard, and will pay off once the CRM, the email platform and the reporting layer are all replaced too.
The test is simple. If you had to stop after this phase and explain why it was money well spent, could you? If the honest answer is “it will be, later,” re-cut the phase.
Decide what you will never automate
Most plans have a list of things to automate. Very few have a list of things that must stay human, and that omission is what makes automation projects feel like they are eroding something.
Write the second list first, and write it down where people can see it. Some candidates:
- Any decision where you would have to explain to someone affected why the answer came out that way.
- Anything where the relationship is the deliverable, such as a thank-you that matters, a difficult conversation, or a first meeting.
- Any judgement you would not delegate to a competent new starter in their first month.
This list does two things. It protects the parts of the work that carry your reputation, and it makes the automation conversation safe to have. People stop defending their whole job when you have already said which parts are staying.
Measure handoffs, not launches
The usual progress metric is whether the new system is live. That tells you almost nothing, because a live system that nobody connected is just a more expensive filing cabinet.
Count handoffs instead. How many manual transfers of information did we have before, and how many do we have now? How long does the most common end-to-end journey take today, measured from the moment information arrives to the moment the work is done? Those numbers move when the project is working and stay flat when it is not, which is exactly what you want from a metric.
What the plan looks like when it is done
Short. A page for the flows, a page for the phases, a page for the never-automate list, and a note on how you will measure. If it is longer than that, it has drifted back into being a software plan.
And review it when a tool changes, not on an annual cycle. The whole advantage of planning the flow is that tool changes are cheap to absorb, but only if you actually absorb them rather than waiting for the next planning season.
Written by
Mike Beasley
Founder, Mike Beasley & Associates LLC
Mike has spent more than fifteen years leading complex business and technology projects, including senior roles in healthcare and education, and now builds automation, donor systems and client portals for nonprofits, associations and small businesses across Virginia, North Carolina and the rest of the US.