Multi-site coordination: synchronising your teams and securing your critical interfaces
Four interfaces concentrate multi-site gaps
A three-week delay on a 6-site office deployment means 40,000 to 120,000 EUR in duplicated rent going up in smoke, not to mention the teams left waiting. Across 34 multi-site operations managed between 2023 and 2024 (3 to 12 locations each), our teams measured that the majority of delays and performance gaps are concentrated on 4 specific inter-site interfaces, not on the sites themselves. It is neither the technical complexity of a site nor the quality of your local teams that causes a deployment to drift: it is what happens between the sites. Kytom handles this orchestration end to end, from mapping local constraints to the weekly coordination committee, with a master schedule featuring local buffers that account for the contingencies specific to each site, in order to limit inter-site delays and stabilise schedule gaps. When ERP constraints govern each location, our method ensures these frameworks do not fragment your deployment. Here is how we proceed, and what the gaps really cost.
The 4 management levers
Four interfaces structure every office deployment we coordinate, and account for the bulk of the drifts observed across our portfolio.
| Interface | What Kytom locks down | Risk avoided |
|---|---|---|
| Schedule-resources | Synchronising the design office, the SPS coordinator (Labour Code R4532), the quantity surveyor across several sites | Unbudgeted travel time |
| Validation-decision | Centralised strategic arbitration, operational decisions delegated to local managers | Bottlenecks of 5 to 8 days per request |
| Supply-delivery | Logistics flows aligned with the actual sequencing (partitions, suspended ceilings, furniture) | Cascading shortages across 3 to 6 sites |
| Quality control-acceptance | Visual standards, acceptance sheets, harmonised finish levels | Untreated ERP/non-ERP gaps |
These 4 interfaces concentrate the bulk of the performance gaps observed across our multi-site deployments. Each one mobilises, at Kytom, a dedicated management tool, a named lead and its own review cadence.
Our stance, against the grain of PMO orthodoxy. Programme literature recommends a breakdown by technical work packages (partition package, HVAC package, furniture package) with a cross-site lead. In practice at Kytom, across 34 operations, this model only works on homogeneous portfolios. As soon as a deployment mixes ERP and non-ERP, occupied and vacant sites, different regions, the breakdown by work package fragments responsibility in the wrong place. The breakdown by interface keeps a single lead on each inter-site value chain and significantly reduces the grey areas.
What you gain
Three mistakes that cost 15 to 25% in delays, and their real price
Three error patterns systematically recur on poorly framed multi-site projects (analysis of 9 operations delivered site by site before switching to centralised coordination).
- Duplicating the pilot site’s schedule without adjusting for accessibility, time slots in an occupied site, local resources or ERP constraints. The theoretical schedule drifts from the second week onwards.
- Centralising all validations at headquarters creates a bottleneck where each local request can wait several days for arbitration, paralysing on-the-ground responsiveness. Across 6 parallel sites, the cumulative effect exceeds 3 weeks of overall delay.
- Pooling specialised resources (acoustician, LV/ELV electrician, technical inspector) without budgeting for travel, remobilisation and getting to grips with the local context.
What these 15 to 25% delays cost your CFO. On a deployment of 6 office sites of 800 m² each (4,800 m² total), a 3-week delay at a headline rent of 280 EUR/m²/year represents roughly 77,000 EUR of duplicated rent (new lease started + old lease still running). In half the cases, this is compounded by a deferral of employee productivity estimated at between 30,000 and 80,000 EUR for 200 to 300 people. The trade-off is clear-cut: around 15,000 EUR of central management over 6 months versus more than 100,000 EUR of losses avoided.
Our approach rests on three levers: a master schedule with local adaptation buffers, explicit delegation of operational validations to site managers, and geographic sequencing of specialised interventions. The result: a significant reduction in inter-site delays, observed across all our recent client portfolios.
When not to go for it
The cases where our framework is not the right answer
Commercial honesty is part of our method. The 4-interface framework is not justified in every context, and we say so upfront.
Below 3 simultaneous sites, or when sites are strictly homogeneous (same landlord, same typology, same time window of 4 weeks maximum), setting up 4 formalised interfaces generates more coordination than gain. A project director and single lead duo is enough. The framework is justified from 3 distinct sites onwards, over a window longer than 8 weeks, with at least two heterogeneous local constraints.
For standardised deployments (the same retail concept or bank branch replicated across 20+ identical sites), duplicating the pilot schedule is, on the contrary, the right strategy, and centralising validations becomes an asset. In these cases, Kytom steers the client towards a classic programme approach by technical work packages, not towards our interface method.
Our interface method is calibrated for heterogeneous office projects: 3 to 12 sites with different local constraints (ERP/non-ERP, occupied/vacant, distinct regions), a deployment window longer than 8 weeks, a requirement for employee productivity from handover onwards. If your operation falls outside this scope, we tell you so at the first meeting and adapt the approach, or even steer you towards the right contact.
Method
- Audit of constraints by site
Our teams take stock, site by site, of accessibility, intervention hours, applicable regulations (Labour Code R4211 to R4217, possible ERP classification) and the local resources available. The aim is to isolate the specifics that cannot be transferred from one site to another before any schedule alignment. This step conditions everything else: a botched audit makes the theoretical schedule drift from the second week onwards. - Master schedule with critical milestones
We identify the milestones common to all sites (partitioning acceptance, HVAC commissioning, furniture delivery) and set local buffers of 10 to 15% per site. The master schedule becomes the single reference shared between the central lead and each local manager. - Mapping of shared flows
We map the resources pooled across sites: specialised teams (acoustician, LV/ELV electrician, technical inspector), long-lead materials, bespoke furniture (3 to 7 variants depending on the project). Logistics circuits are optimised through geographic sequencing to minimise travel and budget for remobilisation time. - Centralised management, local autonomy
We deploy real-time dashboards and automatic alerts on schedule or budget drifts, while keeping operational decision-making as close as possible to the site. Strategic arbitration goes up to the central lead, operational decisions stay with the site managers. This dual rule avoids the 5 to 8 day bottlenecks observed under strict centralisation. - Weekly coordination committee
Each week, the central lead brings together representatives from each site with a standardised agenda: progress by milestone, emerging risks, arbitration to be settled. On a 6-month, 6-site project, this committee represents around 15,000 EUR of central management, to be compared with the 77,000 to 200,000 EUR of losses avoided by keeping delays under control.
Frequently asked questions
From how many sites does centralised coordination become relevant?
From 3 simultaneous sites over a window longer than 8 weeks, with at least two heterogeneous local constraints (ERP/non-ERP, occupied/vacant, different regions). Below that, a project director + single lead duo is enough and formalising the 4 interfaces generates more coordination than gain. Our team tells you so at the first meeting: if your operation does not meet these criteria, we adapt the approach rather than over-scaling the management.
How much does a weekly coordination committee cost on a 6-site project?
Around 15,000 EUR of central management on a 6-month project, weekly committee included, dashboards and automatic alerts included. To be compared with the order of magnitude of losses avoided: 77,000 EUR of duplicated rent for a 3-week delay across 6 sites of 800 m² at 280 EUR/m²/year, plus 30,000 to 80,000 EUR of deferred employee productivity for 200 to 300 people. The economic trade-off is clear-cut, which is why our multi-site office clients make this practice standard.
