'Start small' is advice. Here's how to actually do it

Part of our Q3 series on change management in tax.

The advice everyone gives, and nobody explains

Every tax leader we spoke with this quarter said a version of the same thing. Start small. Pilot it. Learn fast, then scale. Brigitte Baumgartner puts it as low-hanging fruit first. VJ Krishnaswamy calls it fail fast, learn fast, and adopt.

They're describing it from inside their own teams, as the people leading the change. It's the view you need, and it stops one step short of the mechanics: start small where? Which country, how far ahead, and what exactly are you supposed to learn from it?

Those two views complete each other. The leaders in this series have each led this once or twice, from the inside. Sjef de Wit, VAT interim manager and partner at The Resource, has done it many times over, from between the company and the software, in SAP and Oracle environments, has done it many times over, from between the company and the software. So we took what the whole series taught us about change and ran it past his eye for what a rollout actually costs you.

First, know which kind of rollout you're in

He draws a line before anything else, and it's worth stating up front. With a full system replacement you often have no choice but to go big bang. With a tax tool you do have a choice, and he's unambiguous about which one to take.

"With a rollout, I think it’s very important that you work in steps."

Big bang on a tool rollout tends to cost you more than it saves, because the rework arrives everywhere at once. If you're replacing an ERP, this article isn't for you. If you're rolling out a tax platform, keep reading.

Pick the country that carries every flavor

The pilot country is the decision that determines everything after it, and most teams pick it on the wrong criteria. Not the smallest. Not the biggest. One of the larger ones. Not the hardest jurisdiction either: "you don’t start with Italy or Poland," he says. Countries with heavy local requirements teach you about the country, not about your rollout.

The real test is different. The country you pick has to contain every version of your business. If you have production somewhere, a sales office somewhere else, and a few loose sales agents in a third place, then "you want a country where all three flavors occur." Work out which flavors exist in your business first, then find the country that carries them. And choose it with your team rather than for them.

This sits alongside VJ's advice in Module 4 rather than against it. He deliberately mixed material and lower-risk entities instead of saving the hard cases for last, which is about not deferring risk across a whole program. This is about which single country goes first. Both are right, at different altitudes.

Let it run two to three weeks ahead

Here's the part that turns phasing from a principle into a schedule.

"It can be very close together. The first country can run two or three weeks ahead."

Not months. Weeks. And the point of the gap isn't caution, it's arithmetic:

"It’s very annoying if I make that mistake in the Netherlands and at the same moment in Portugal and in France. Then I have to go back to AP for all of them and pull all that data again."

One wave ahead means you find the awkward parts of your data once, fix them once, and arrive at wave two already knowing what will be asked of you. Starting small doesn't mean going slow, because you can learn fast.

Worth being precise here, because Module 4 also tells you a full tax cycle runs twelve to eighteen months before you can really judge a change. That's the time to a verdict. This is the gap between waves. Different clocks.

Your first wave maps your stakeholders for you

"People are averse to text."

So don't write a plan, draw one. Week one, week two, week three, week four: this is what we do, this goes live, this doesn't. Color it red and green. Where are we ahead, where are we behind, and what are we doing about it. Then walk it with your key person every single week.

If that sounds like the watermelon effect VJ warns about in Module 4, green on the surface and red underneath, the difference is what happens next. A color is only worth something if someone acts on the red in the same week that it turns red. Which is the next section.

Settle one thing in week one, too: how your sponsor wants to work. Some want every email in copy. Others want a weekly catch-up and a plan they can read. Your whole rhythm depends on which one you're dealing with.

Escalate the same week

This is where the sponsorship you won in Module 2 finally gets spent. The moment a department goes quiet, it goes up, immediately, with names attached. His own example:

"Dear tax director, you want Keeyns live, but Treasury won’t free up a single moment for me this month. I’m about to get stuck."

Not a complaint. A fact the sponsor can act on. And if you're the tax director, that message is your job, not an interruption of it.

He's clear that outright resistance is rarer than people expect. It was decided above, and plenty of colleagues have a stake in it working. What earns you cooperation is taking work off their hands: all of this has to be delivered, I want to do as much of it for you as I can, and if you'd rather do it yourself, that's fine too.

Finish before you add

One clean rule, and it saves projects. Follow-on scope starts when the first phase is properly done, not while it's still running. Every phase you leave half-finished becomes rework you pay for twice.

The honest part

The last thing he told us is the thing most vendors leave out, and he says it about us too:

"Nothing is plug and play. I buy something, plug it in, and it works. That’s my bread and butter. Nothing is plug and play."

"A lot has to be put in. If you don’t put time and effort into it, nothing comes of it."

A tax platform is an order of magnitude lighter than an ERP program, and it still needs your data, your decisions, and someone's time. A buyer who was promised smooth and then meets week three feels misled. A buyer who was told the truth arrives at week three on schedule.

That's what a pilot is really for: to find out what this costs you, once, in one country, while it's still cheap to find out.

Keeyns gives you one place to see every entity, process, owner, and deadline, so the plan you run the rollout on is the same plan you run tax on afterwards. Stichd runs VAT compliance across more than fifty markets with two people, and hasn't missed a deadline since go-live.

This article is part of our Q3 series on change management in tax. Explore the full series, including our change management crash course, to see how leading teams are getting back in control.