How to Design an Activity-Based Office: A Framework

Activity-based working (ABW) gets pitched in India as a way to cram more people into less carpet area, and that is exactly why so many fit-outs disappoint. A floor full of beanbags and phone booths that nobody booked, while the same teams fight over the same six desks every morning, is not ABW. It is a desk shortage with better photography. Done properly, activity-based office design starts from how your people actually work, assigns a mix of settings to those patterns, and proves the mix on your real floorplate before a single partition is ordered.

This is a step-by-step framework we use with enterprises and GCCs across Mumbai and Pune. It assumes you are not greenfield. You have an existing lease in BKC or Hinjewadi, a headcount number that keeps changing, and a hybrid attendance pattern nobody fully trusts. The goal is a defensible plan: ratios you can justify to finance, a layout you have test-fitted, and a pilot that de-risks the rollout before you spend on Cat-B.

Step 1: Run a real utilisation audit

Before anyone draws a layout, measure how the current floor is used. The number that matters is peak occupancy, not assigned headcount. In most Indian hybrid offices we audit, badge data shows peak attendance running at 55–70% of seats on the busiest day, and desk-level sensing shows assigned desks sitting empty 40–60% of the working day even when their owner is in the building. Because that person is in a meeting room, at a colleague's desk, or in the cafe.

Gather three data layers and overlay them:

• Badge/turnstile data for true daily and hourly attendance across at least 8–10 weeks, so you catch the Tuesday-to-Thursday hump and month-end spikes.
• Meeting-room booking + actual sensor data. The gap between booked and occupied tells you how much of your room inventory is being hoarded. Large rooms booked for two-person calls is the classic finding.
• A point-in-time observation study. A walkthrough every 60–90 minutes for two weeks, logging what each setting is being used for. This is the only layer that captures behaviour: heads-down work happening in collab zones, calls taken in stairwells, focus rooms used as storage.

The audit output is a single utilisation curve and a setting-by-setting occupancy table. That table is your evidence base for every ratio decision that follows, and it is what you will show finance when they ask why you are not simply giving everyone a desk.

Step 2: Map personas and work patterns

Headcount tells you how many people. Personas tell you what they need. Group your population not by department but by work pattern, because a developer and a financial analyst may sit in different orgs yet have nearly identical needs (long heads-down stretches, occasional pairing, rare large meetings).

A workable set of personas for a typical GCC or enterprise floor:

Size each persona as a percentage of headcount using HR data plus the badge curve, then validate with short manager interviews. They will tell you which teams genuinely cannot hot-desk (trading, certain regulated functions, hardware labs) and must keep fixed neighbourhoods. This persona mix is the spine of your workplace strategy; it determines the entire downstream setting count.

Step 3: Set the setting mix and ratios

Now convert personas into a setting schedule. Two ratios drive the design: the sharing ratio (people to workpoints) and the collaboration ratio (the split between individual and shared/meeting settings).

Sensible starting points for an Indian hybrid floor, to be tuned by your audit:

• Sharing ratio of 1.2:1 to 1.5:1 for flex-heavy populations. I.e. 100 flex staff share roughly 70–85 workpoints. Anchor-heavy teams stay closer to 1:1.
• Individual-to-collaborative split moving from the old 80:20 toward 60:40 or 55:45 as attendance concentrates on collaboration days.
• One enclosed focus/phone room per 8–12 people. Chronically undersupplied in Indian offices and the single most common post-occupancy complaint.
• Meeting rooms re-sized down: more 2–4 person rooms, fewer 10+ boardrooms, because the audit will show large rooms are your worst-utilised asset.

Be deliberate about acoustics and air: a denser, more varied floor only works if MEP keeps up. Diversity zoning (quiet vs. Buzz) has to align with your HVAC zoning and acoustic treatment, which is why setting mix should be agreed with your MEP engineering team before the layout freezes, not after. Build the schedule into the BOQ as a counted list. X focus rooms, Y huddles, Z touchdowns. So it survives value engineering with its logic intact.

Step 4: Test-fit on your actual floorplate

A setting schedule that works in a spreadsheet can fail against a real core, a single fire stair, or a column grid that won't take your desk cluster. The test-fit is where you prove the ratios fit the building. Take your setting schedule and block it onto the actual floor plate, respecting the structural grid, the position of cores and toilets, daylight, and. Critically. Fire and egress compliance for the higher densities ABW implies.

A good test-fit answers concrete questions: Does the target headcount actually fit at the chosen sharing ratio once circulation (typically 25–35% of carpet area) is honoured? Where do the buzz zones land relative to the focus rooms? Can you create distinct neighbourhoods, or does the core chop the floor into awkward fragments? We usually produce two or three test-fit options per floor and pressure-test each against the persona counts and the fire NOC implications before recommending one. Seeing it in 3D and walkthroughs at this stage catches problems. A phone booth opening onto a quiet zone, a collab table under a return-air grille. That plans alone hide.

Step 5: Prove it with a pilot zone

Do not roll ABW across three floors on day one. Convert one neighbourhood. 40 to 80 seats for a representative team. Into a live pilot, built to near-final spec, and run it for 6–10 weeks. The pilot exists to find what the audit and test-fit could not predict: whether people actually book the focus rooms, whether the booking tool gets used or ignored, whether the sharing ratio holds on a Wednesday, where the noise complaints land.

Treat it as an experiment with before-and-after measurement. Instrument the pilot zone with the same sensing you used in the audit, survey the team at week one and week eight, and let the findings rewrite the ratios. It is far cheaper to discover you need one more focus room per cluster in a pilot than across a completed turnkey fit-out. The pilot also becomes your internal proof case. Leaders who walk it tend to stop arguing about assigned desks.

Step 6: Enable, manage the change, and measure

The physical setting mix is only half of ABW. Three things make it stick:

• Tech enablement. Desk and room booking, wayfinding to free settings, and occupancy sensing that feeds a live dashboard. Lockers for a clean-desk policy are non-negotiable. Without personal storage, hot-desking quietly collapses back into territorial desks.
• Change management. ABW is a behaviour change, not a furniture change. Publish a simple neighbourhood etiquette (clean desk, booking norms, call-room courtesy), train floor champions per team, and have leadership visibly use the shared settings. In Indian offices, where the assigned desk often signals seniority, leaders sitting in the open is the strongest possible signal.
• Measurement. Define success metrics before move-in: peak utilisation against the sharing ratio, focus-room booking rates, employee satisfaction, and cost per seat. Re-measure at 30, 90, and 180 days and adjust the mix. ABW is meant to be tuned, not frozen.

If you align these with how you brief and budget the project, the workplace strategy, the test-fit, and the build all reference the same persona and ratio decisions. Which is what keeps the design honest from concept through DLP.

Frequently asked questions