Delivery & Steering

Managing Collaboration Across Locations

Marin Niehues25 May 20269 min read

A team reviewing drafts on a pinboard

Collaboration across locations succeeds when each site has clear goals and decision rights and handovers are bindingly defined. Distance itself causes less friction than vagueness at the interfaces. Anyone leading several locations must therefore deliberately organize what happens by itself in the corridor at a single site. This article presents a control model with four levers, an interface check per pair of sites and the metrics by which executive management and business-unit heads can read the effect.

Why is collaboration across locations difficult?

Collaboration across locations is difficult because distance removes the casual exchange through which teams at one site clarify questions and spot misunderstandings early. Research has described this effect with remarkable consistency for almost fifty years.

In the 1970s, Thomas Allen studied communication in research and development organizations at MIT and published his findings in 1977 in “Managing the Flow of Technology”. The frequency of technical communication between two people falls rapidly as physical distance grows. From around 30 meters apart, or one floor's difference, communication drops almost to the level between separate buildings. This pattern is known as the Allen curve.

Between cities and time zones, casual exchange largely disappears, and what used to be settled in passing now needs a meeting. In their work “Distance Matters” from 2000, Gary and Judith Olson showed that video conferencing and shared tools reduce distance but do not fully replace proximity. Distributed work succeeds best when tasks are clearly divided and the teams share a common base of knowledge and working practice.

In 2001, Catherine Cramton described the mutual knowledge problem: distributed teams know little about the context of other sites, for instance their priorities or bottlenecks. In 2003, Pamela Hinds and Diane Bailey set out why distributed teams are more prone to conflict: shared context is missing, and misunderstandings surface later than at a single site.

For delivery, the consequence is clear. Work stalls at handovers, where one site passes results to the next and nobody sees the gap. An organization with several locations therefore steers above all the quality of its interfaces.

How do you steer collaboration across several locations?

You steer collaboration across several locations through four levers: goals, decision rights, handovers and information flow. Each lever can be assessed per pair of sites and assigned to a named owner.

  • Goals: Each site knows its contribution to the overall result and the goals of neighbouring sites. Goal conflicts are named before they surface in execution and decided in a joint meeting of the site leaders.
  • Decision rights: For every recurring decision it is defined where it is taken and by when. Without this rule, decisions migrate to the largest site or to headquarters, and waiting times grow.
  • Handovers: Every handover between sites has a format, a point in time and a recipient who confirms it. Time zones become capacity when work moves predictably from site to site.
  • Information flow: There is a defined place where status and decisions are documented, and a fixed rhythm in which all sites see the same status.

The levers build on one another. Goals and decision rights define who is responsible for what, handovers turn that into work, and information flow makes both visible. How decision rights can be clarified with RACI and RAPID is covered in Clarifying decision rights.

A second rule concerns structure. In 1968, Melvin Conway observed that organizations produce designs that mirror their own communication structure. Where sites are loosely connected, this is reflected in products and processes. Divide responsibility so that tightly coupled work sits at one site and only interfaces with a clearly defined output run between sites.

The interface check per pair of sites

The rules a pair of sites needs follow from a few criteria. The checklist below sets, for each criterion, the threshold above which a stricter rule applies.

The thresholds are starting values that can be sharpened during testing. What matters is that every pair of sites has a deliberate rule and that the metrics show whether it works.

Where collaboration across locations breaks down in practice

Delivery shows recurring patterns. The first is the information advantage of the main site: decisions are taken where leadership sits, and the other sites learn of them late. The second is the duplication of capabilities that arises when several sites build the same thing because nobody clarifies ownership. The third is the handover without confirmation of receipt, where work stalls without anyone noticing.

The countermeasures follow from the levers. Decision rights are allocated per decision type, each capability gets a lead site, and handovers receive a confirmation and a deadline. The related question of how functions collaborate with one another is covered in Breaking down silos.

Which metrics show whether steering is working?

Four metrics, set at the interfaces and collected per pair of sites, show whether steering is working:

  • Waiting time for decisions between sites
  • Lead time per handover
  • Queries and rework after a handover
  • On-time delivery for work that involves several sites

These values reveal earlier than sentiment surveys where collaboration is stalling. They are measured per pair of sites so that the cause of a delay can be attributed to an interface. A pair whose waiting time rises over two quarters goes on the agenda of the next steering meeting. An overview of delivery metrics is given in Measuring delivery capability.

Niehues Consulting as a partner for multi-site undertakings

Niehues Consulting steers undertakings across locations as delivery manager, project manager or product owner and designs the structures for them. Founder Marin Niehues works as a management consultant from the delivery perspective: he builds organizations that deliver and brings AI into operations. His projects range from VC-funded scale-ups to DAX 40 groups, with assignments in 20+ countries on six continents and a home base in the DACH region.

The approach follows the Niehues Delivery Model with the phases Understand, Build and Implement. In Understand, the team establishes which value streams touch several sites and where the handovers lie. The guiding test questions are: who decides, who delivers, who uses? What is the baseline, and how do we measure success? The result is a map of the interfaces with today's waiting times.

In Build, decision rights, handover formats and the rhythm of coordination are designed for one value chain and tested there. The test question of this phase: which roles, decision paths and interfaces does operations need? One chain is enough to prove the effect. In Implement, the model is transferred to further chains, with responsibilities and enablement of the site leaders. How organizational design and delivery management interlock is described in the article characteristics and metrics of a high-performing organization.

The entry point is the free initial consultation of 30 minutes, in which you can assess your site landscape.

Frequently asked questions

What is the Allen curve?

The Allen curve describes how the frequency of communication between two people falls sharply as physical distance increases. Thomas Allen established the pattern in the 1970s at MIT in research and development organizations. From around 30 meters apart, or one floor, exchange falls almost to the level between separate buildings. For multi-site organizations, coordination therefore has to be organized deliberately.

Which tools are suitable for collaboration across locations?

Suitable tools for collaboration across locations are a shared document repository, a task system with visible status for each handover and a decision log that all sites can see. Chat and video complement this basic set for quick coordination. The tools become effective only with firm rules on where decisions are documented and who confirms handovers.

What is the biggest obstacle with distributed sites?

The biggest obstacle is the lack of shared context. It leads to delayed decisions and to misunderstandings at handovers that surface only when deadlines slip. Documented decisions and regular visits between sites counter it.

Do all sites need the same processes?

All sites need the same interfaces; their internal processes may differ. Work within a site can be organized differently as long as handover format, deadlines and decision rights apply. Standardizing internal processes pays off once employees move frequently between sites.

What is Conway's law?

Conway's law states that organizations design systems that mirror their own communication structure. Melvin Conway formulated it in 1968 in the magazine Datamation. For multi-site organizations this means that tightly coupled work belongs at one site, with clearly defined interfaces between sites.

How should international teams handle time zones?

International teams handle time zones best by reserving shared working hours for decisions and handing over everything else in writing. With less than three hours of overlap, handover with confirmation becomes the standard. Steering across countries is covered in more depth in The international organization.

How do you build trust between sites?

Trust between sites grows through commitments kept at handovers and through personal contact at the start of a collaboration. Teams that reliably meet deadlines and document decisions transparently build trust faster than teams with many coordination meetings. A joint kick-off in person pays off for every new combination of sites.

Let’s talk about your initiative.

In 30 minutes we clarify where your organization stands and which step will pay off most now.

Request a consultation

30 minutes · free · no obligation