The core/flex ratio for Series B-C scale-ups: stop guessing 70/30

Core space = traditional commercial lease, your name on the door, you control the layout, multi-year commitment. Cost per m² is lower (€500-800/m²/year in Paris central vs €1,000-1,500 equivalent for flex), but you pay separately for fit-out (€500-1,500/m² one-time), services, IT, utilities, taxes, insurance.

Flex space = operated office (Wojo, Deskeo, WeWork, etc.), 6-24 month commitment, all-in pricing, plug-and-play. Higher per-desk cost but no upfront capex, faster setup, easier exit.

The 'right' ratio is the split between these two for your team — what fraction of headcount sits in core vs flex, and what fraction of the workspace is core-leased vs flex-rented.

Most CFOs conflate the two questions. You can have 100% of headcount sitting in flex (small team, high uncertainty) or 100% in core (large stable team, low uncertainty). The 70/30 default is rarely optimal for either extreme.

Before discussing ratios, we ask: what's your headcount confidence interval at 12 months? Specifically, what's your P10, P50, P90 forecast for total team size in 12 months?

If P10 = 25, P50 = 45, P90 = 80 — you have huge uncertainty (3.2x range from low to high). Solution: lease for P10 in core (locked-in cost efficiency for the certain headcount), put everything above P10 in flex (you can scale up to P90 or scale back if you miss P50).

If P10 = 60, P50 = 65, P90 = 75 — you have very low uncertainty. Solution: lease for P50 in core, put a small flex buffer (maybe 10%) for unexpected hires.

The ratio falls out of the math, not the other way around.

Pattern A — High-velocity Series B (50→120 in 18 months): Core 30-40%, flex 60-70%. The core is for the stable engineering / ops team. The flex absorbs the GTM expansion. We've seen this pattern compress total real-estate cost by 18-25% vs a single core lease for the P50 headcount.

Pattern B — Series C consolidating after fast growth (200→230 in 12 months): Core 70-85%, flex 15-30%. The team is now stable enough to commit to a 6-9-year traditional lease. Flex remains for satellite teams (regional sales, contractor pods).

Pattern C — Distributed/hybrid mode (40 employees, 3 cities): Core 0-20%, flex 80-100%. Doesn't make sense to commit to traditional leases when no city has critical mass. Flex everywhere, single contract via Wojo or WeWork for cost efficiency.

The ratio that's almost always wrong: 70/30 by default with no math behind it. We've seen that split applied to all three patterns above and it underperforms in all three.

Two scenarios where the core/flex framework fails:

Scenario 1: hybrid policy is in flux. If you're deciding between full-remote, 2-day-office hybrid, and 4-day-office hybrid — the headcount in office on a Tuesday vs Thursday differs by 30-50%. The ratio depends on how you size the office: peak day or average day. We've seen firms over-lease core because they sized for peak, then sit at 50% occupancy on Mondays/Fridays. Flex would have been the right call.

Scenario 2: real-estate-as-recruiting. If the office is a hiring tool (premium location, beautiful interior, in-person culture differentiator), the ratio matters less than the address. We've seen Series C firms commit to a Deskeo flagship at 100% flex because the conversion uplift on senior hires justified the cost — they could have done a cheaper traditional lease elsewhere, but it would have hurt recruiting.