Insights / Guide / Operating model design
GuideOperating model design: the complete guide
The operating model is how strategy becomes daily work. How to assess yours, choose a structure, set decision rights and measures, and manage the transition.
The operating model is how an organisation turns strategy into daily work: its structure, roles, decisions, processes and measures. When it fits, people know what they own, decisions are made at the right level and work flows across teams without constant escalation. When it does not, even a good strategy stalls somewhere between the boardroom and the front line.
This guide brings together the briefings in our operating model series. It explains how to assess the model you have, the main design choices and their trade-offs, and how to move to a new model without losing the confidence of the people who must run it.
Why strategies stall between boardroom and front line
Four patterns recur when strategies lose momentum. First, too many priorities: when everything is important, teams default to what they were already doing. Second, no owner: a goal that belongs to a committee belongs to nobody. Third, a budget that still follows last year’s structure, so money and people never move. Fourth, measures that describe activity, such as meetings held or reports produced, rather than results.
Each of these is an operating model problem rather than a strategy problem. Structure, decision rights and governance decide how quickly strategy turns into action. If a decision needs six approvals, the strategy moves at the speed of the slowest approver. Designing who decides what, and how fast, is often worth more than redrawing the organisation chart.
A simple diagnostic
Write your strategy on one page. Ask five managers from different levels to explain what it changes for their team. Where their answers diverge is where the operating model is failing to translate strategy into work.
Assess how you operate today
Start by mapping the current model honestly: structure and spans of control, decision rights, core processes, systems and the forums where the business is run. Interviews tell you how people experience the model; data on approval times, escalations and duplicated roles tells you what it actually costs.
- Structure: units, reporting lines, layers and spans of control.
- Decisions: who really decides on pricing, hiring, spending and customer exceptions.
- Processes: where work crosses teams and where it waits.
- Governance: which meetings exist, who attends and what they actually decide.
- Measures: which reports are used to run the business, and which are produced but ignored.
Look for signs that the model is working against the strategy: duplicated functions across business units, managers with one or two direct reports, decisions routinely escalated to the chief executive, or support teams that the business has stopped trusting.
Be careful not to treat the assessment as a search for culprits. Most operating model problems are the result of sensible decisions made at different times for different reasons: a role created to cover a crisis, a committee set up for a project that ended, a reporting line drawn around a person who has since left. Framing the work as fitting the model to today’s strategy, rather than fixing mistakes, keeps people open and honest about what they see.
Choose the structure and design the roles
No structure is perfect. Each optimises something and costs something else, and the right choice depends on what the strategy most needs.
- Functional structures build deep expertise and efficiency, but can slow work that crosses functions.
- Divisional structures, by product, customer or region, bring focus and accountability, but duplicate support functions.
- Matrix structures balance both, but demand clear rules for who decides what, or they produce endless negotiation.
Whatever the structure, write role charters for leadership roles: what the role owns, what it decides, what it is measured on and how it works with neighbouring roles. People need to see their place in the new model. Role clarity and fair transitions decide whether the model is accepted or quietly resisted.
Shared services: when centralising helps
Centralising common work such as finance, procurement or IT support can save cost and improve control, but only with discipline. Standardise the process before moving it, because centralising a messy process spreads the mess. Define service levels so internal customers know what to expect and how to raise issues. And keep the shared team connected to the business, or it will slowly lose its trust.
Set decision rights and governance
Decision rights are the real design. Many organisations are slow not because people work slowly but because it is unclear who may decide. Slow or unclear decisions are the most common hidden cost in an operating model, and among the cheapest to fix.
- List the ten decisions that happen most often or cost most when delayed.
- For each, record who decides, who advises, who must approve and who must be informed.
- Remove approval layers that do not reduce real risk; each layer adds days.
- Give people clear financial and risk limits so they can decide without asking.
- Set the forums that run the business, with a clear purpose and decision scope for each.
Governance forums deserve the same scrutiny. A typical mid-sized group might find a dozen recurring leadership meetings, several covering the same ground and few with the authority to decide. Fewer, better-defined forums save senior time and make it obvious where each decision gets made.
Define performance measures leaders use
A good performance framework has few measures that people understand and can influence. Each measure should trace back to a strategic priority; if it does not, question why it exists. Every measure needs an owner and a regular review slot. If nobody has used a report in a quarter, stop producing it.
Measures drive behaviour, so choose them carefully. A sales team measured only on revenue will discount to win; one measured on margin as well will make better trade-offs. Test each proposed measure by asking what behaviour it will encourage at the edges.
Keep the set small. Five to eight measures per leadership area are usually enough to run it well, with more detailed operational measures held at team level where they are actually used.
Plan the transition
Design is the easier half. Moving an organisation to a new model affects jobs, reporting lines and habits, and people will judge the model by how the transition is handled. Sequence the move carefully, with clear communications, an honest account of people impact and a plan for the first hundred days.
We spent two months on the design and should have spent three on the transition. The structure was right; it was the uncertainty in between that cost us good people.
Group HR director, distribution client
There is a trade-off between speed and disruption. A single, well-prepared switch limits the period of uncertainty but concentrates risk. A phased move lets you learn and adjust but prolongs ambiguity. Many organisations change decision rights and governance first, because that delivers quick benefits with little disruption, and move structure once the new ways of deciding are working.
A well-designed model delivers clear ownership of every process and decision, faster decisions with fewer escalations, lower cost to serve through less duplication, and a structure ready for new entities, markets and systems. A practical first step is the decision list above: map your ten most frequent or costly decisions this month, and see how many approvals each really needs.
Source: Intelliversal analysis, drawing on the briefings in this series.
