Standardize the Platform, Localize the Experience

I spent 18 years running global workplace services across 180 sites in 20 markets, and for most of that time the mandate was standardization. Same standards, same programs, same playbook, everywhere. We executed it with real discipline too. This wasn’t a half-hearted corporate initiative that fizzled out. It worked, in the sense that the program did what it was designed to do.

Compliance was strong, the dashboards stayed green, however in a lot of our markets the experience quietly got worse. We were measuring inputs while the outcome eroded, and it took me longer than I’d like to admit to see it clearly.

What follows is the model I’d build instead. I want to be upfront that this isn’t the model we ran. It’s what I concluded after watching the pure version play out at scale, including its eventual half-correction. The premise is simple: workplace experience is an outcome, not a standard, and the same inputs produce completely different outcomes depending on where people are and what they already have.

The Maslow problem

Maslow published his hierarchy of needs in 1943. It might be the most useful diagnostic tool in global workplace strategy, and probably the least used.

The idea is that needs are sequential. You can’t deliver esteem to someone whose safety needs aren’t met. And you can’t impress someone with safety they already take for granted.

Now apply that to a global portfolio. In some of our markets, great workplace experience started at the bottom of the pyramid. Backup power that actually held through an outage, not on paper. Safe transport home for people working after dark. Security at the door that protected without intimidating. Air you could trust. Get any of those wrong and nothing else you spend money on registers. The coffee program means nothing to someone who’s worried about how she’s getting home tonight.

In other markets, all of that is invisible. Nobody in Geneva thanks you for uninterrupted electricity. There, the fight is higher up the pyramid: natural light, acoustics, spaces that match how people actually work, and these days, an office that can justify its own commute.

Here’s the trap, and I fell into it along with everyone else. Standardization programs get written at headquarters, and headquarters lives at the top of the pyramid. So the program mandates the amenity package, the service levels, the vendor list, and ships it globally under the banner of consistency. What you actually get is esteem-level spending in markets still fighting at the safety level, and redundant spending in markets that needed something else entirely. Consistent inputs. Wildly inconsistent experience.

The same thing happens inside a single building. An engineer who needs three hours of quiet, a team that lives at the whiteboard, a new hire trying to find her footing, a parent who needs the day to end on time. One office, four different definitions of a good day. Treat them identically and you serve none of them particularly well.

Why smart companies do it anyway

I don’t think this happens because leadership is careless. It happens because of a measurement asymmetry that’s genuinely hard to fight.

Procurement savings from standardization are precise, immediate, and easy to put on a slide. Experience erosion is diffuse and delayed, and it shows up in numbers nobody connects back to the workplace: attrition, disengagement, offices that slowly empty out. So in every budget argument, the measurable number beats the real one. I sat in those meetings. The side arguing for the spreadsheet always had better ammunition than the side arguing for something people could feel but nobody could prove.

What happens next is the part I find most telling. When the erosion finally becomes undeniable, companies retreat, but watch which layer they retreat on. It’s always the cheapest one: visual identity. Local branding flexibility gets granted, the offices get some local color, and leadership declares the balance restored. The layer that actually determines experience, the services, never moves, because touching services would threaten the savings math that justified the whole program. Where service variation survived at all in my experience, it survived out of necessity. A transport program you simply couldn’t cut in a market where people had no safe way home. Nobody designed that. It just refused to die. Cosmetic flexibility isn’t localization. It’s an apology.

The three-layer model

The answer isn’t infinite customization either. A hundred offices each doing their own thing collapses into chaos, cost, and a brand that means nothing. The real question is deciding, deliberately, what must be identical everywhere, what must never be, and who gets to decide. My answer has three layers.

Layer 1: Standardize the platform

Most of the workplace should be fixed, and shouldn’t be up for local debate at all.

This is the kit of parts: the space types every office needs and the ratios between them. It’s the non-negotiables that protect the floor of the experience: indoor air quality, acoustic standards, furniture quality, accessibility, safety systems. It’s the technology layer, which I’d treat as absolute. Audio and video should work identically in every room in every country, so a call between two offices never fails and someone relocating between sites is productive within an hour of walking in.

It’s also the baseline services that keep that floor intact day after day: cleaning frequencies, security requirements, maintenance response times, how the front of house runs. These are classic hygiene factors. Nobody notices them when they’re right, and everybody notices when they’re wrong. They belong in the platform precisely so that no site can trade them away for something more visible. A site that cuts cleaning to pay for a coffee bar hasn’t localized anything. It’s broken the floor.

What the platform shouldn’t cover is anything cultural, anything aesthetic, anything whose value depends on where you are. Standardization genuinely earns its keep here, and in my view this is the only place it should live.

Layer 2: Localize the identity

A small share of the budget makes the office belong to its city, and it buys far more emotion per dollar than almost anything else in the fit-out.

Materials, art, color, the story the space tells about where it is. An office in Hyderabad shouldn’t feel like a photocopy of one in California. When it does, the message people receive, whether anyone intends it or not, is that their market is an outpost.

This is also the layer companies concede when standardization backfires, precisely because it’s cheap. It matters. It’s just not enough on its own. Identity without the next layer is decoration.

Layer 3: Flex the services by value, not by habit

This is the layer most global programs never build, and it’s where the experience is actually won or lost.

The brief to a site should never be “copy HQ.” The brief should be a question: what does this population lack that money can fix?

Transport is one of the cleanest examples. Next to a metro station in a Western capital, a shuttle program adds almost nothing. In a developing city it can be a lifesaver, and I don’t mean that figuratively. It decides whether women on your team can work late, whether your office draws talent from across the city or just one neighborhood, and whether people stay. Same line item, opposite value.

Food works the same way. Free meals are decisive in markets where there’s nothing decent to eat near the office, or commutes are brutal, or meals are simply a competitive hiring norm. In other markets the same program goes almost unnoticed. A barista bar is a genuine gathering point in one coffee culture and an expensive machine nobody touches in another.

So the discipline of this layer is a ranking exercise, done market by market. List the candidate services. Score them against what this population actually lacks and where the company is headed. Fund from the top down. And be willing to fund nothing in a category the market already provides. If the office is surrounded by forty lunch spots, the food line can be zero, and that’s a correct decision, not a stingy one. The moment every site gets something in every category, you’ve rebuilt standardization through the back door, just with local logos on it.

One thing to keep honest here: the proportions. The platform, including its baseline services, is the overwhelming majority of the spend, and it stays locked. The flex is a small envelope at the margin. But that small envelope is the difference between an office that’s merely compliant and one that’s genuinely good for its specific population. The platform is what people expect. The flex is what they feel as care.

Governance, or how the flex stays coherent

Local flexibility without governance turns into twenty different companies. Three rules keep it from getting there.

First, the outcomes are fixed globally. Every site is accountable to the same things: people are safe, people feel the company invested in them, people can do their best work, people are proud to bring someone through the door. Sites vary in how they get there. They never vary in what they’re accountable for.

Second, the flex is a budgeted envelope, not an open tap. The site proposes the allocation, and a global function challenges it against the outcome standards. Local judgment inside a global frame. This is exactly the piece that was missing where I watched standardization fail: when flexibility finally arrived, it was granted but never funded and never governed, which made it decoration.

Third, measure outcomes, not inputs. Dashboards that track compliance and vendor performance will stay green while the experience dies.

And I’d be careful about assuming surveys are the escape, because surveys get managed. Anyone who’s spent time in this function has seen it: the events that cluster suspiciously close to survey season, the showcase of the year’s work, the freebies and the pizza, and then the survey goes out while goodwill is warm. I don’t even think it’s dishonest, exactly. Make a team’s performance appraisal depend on an annual sentiment snapshot and they’ll optimize the snapshot. The score goes up, the experience hasn’t moved, and leadership ends up with something worse than no data, because now it’s confidently wrong.

So weight what people do over what they say. Does the office win the commute in ordinary weeks, not event weeks? Do people stay? Do issues stay fixed after they’re closed, or do they reopen? Do people use the space for things they’re not required to attend? Behavior doesn’t fill out a survey to be polite.

The short version

Standardization is a method, not a goal. The goal is an experience that lets people do their best work, and the method quietly becoming the goal is how that dies while every dashboard stays green.

Standardize the platform. Localize the identity. Flex the services around what each population actually lacks. Hold every site to the same outcomes, and give them real freedom, with real money behind it, on the inputs.

That’s how one company lives well in twenty different realities.