Real Experiences

Late testing doesn’t need heroics — it needs a better operating model.

In large integrations, Business Acceptance Testing (BAT) is supposed to be the moment the business proves the system works for them.

But in Wave 8 of the Gillette SAP APO DRP integration, BAT didn’t fail because the system was broken.

It stalled because the business wasn’t ready to execute.

The Gillette integration was an eight-wave program. I served as the IT technical lead for the first wave (Latin American blades and razors) and the last wave (North American Oral-B and Braun).

Wave 8 reached BAT with low buy-in and delayed engagement—configuration inputs lagged, testing scenarios lagged, and training lagged.

And within the first few weeks, it became clear: the business team wasn’t going to finish BAT on schedule.

VMware wasn’t the hard part — making the work repeatable was.

Virtualization was viewed as a major cost-savings opportunity.

But when I took over the program, the constraint wasn’t the platform.

The constraint was repeatability.

Each site proposal was essentially a custom job: custom equipment scope, custom pricing, and custom analysis to estimate savings. Even basic baseline questions—like what a site was paying for its current infrastructure—weren’t easy to answer consistently.

Capital and construction don’t make you ready. Qualification does.

This is the story of a contract manufacturer → internal site migration where the dramatic part was getting equipment into the building… and the deciding part was earning the right to ship.

In operations, we love things you can see.

Construction. Installs. Equipment arrival.

That feels like progress.

Readiness doesn’t always look like progress.

Until you’re in PQ and the process won’t cooperate.

A stage gate isn’t real until it lives in the system.

Stage gates are one of those ideas that everyone agrees with.

They sound like discipline.

They sound like predictability.

They sound like “how serious teams run projects.”

But there’s a version of stage gates that never changes anything.

It’s the version that lives in a slide deck.

And there’s the version that actually improves execution:

The one that turns each gate into clear milestones, assigns owners and due dates, and makes progress visible in the tool where work happens.

Most plans underestimate artwork.

This is the story of how a capacity ceiling + a missing PO path could have quietly stalled delivery—and what I did to prevent that.

In operations, we love things you can see.

Equipment. Construction. Line trials.

Those feel like “real progress.”

Artwork feels… administrative.

Access Before Action.

Some projects don’t stall because the work is hard.

They stall because the people can’t even get in.

I was brought in on a short-term contract to support a global pharma company on time-sensitive clinical-trial documentation feeding an FDA submission.

We had the team. We had the urgency. We had the assignment.

What we didn’t have was a clean path into their systems.

Ambiguity slows teams. Ownership restores speed.

Systems don’t break because people don’t care.

They break when ownership is unclear.

During a contract manufacturer onboarding at Thriving Brands, I got close enough to the day-to-day work to spot a problem that looked “small”… until it wasn’t:

Our SKU descriptions weren’t consistent.
Not a little inconsistent.
All over the place.

Pressure doesn’t create problems—it reveals them.

The cease-and-desist hit early, and we had to build the artwork system while the work kept moving.