Operating Models Designed Right
For many, operating models have a bad name.
Consultants come in, design a perfect model and then nothing quite goes right when it comes to implementation. There are the classic failings… a feeling that you are re just moving deck chairs; relaunching old meetings and controls that were dismissed two or three op models ago or putting in processes that people have not bought into. The list is endless and why some leaders avoid this type of work, or if they do undertake it, being sure to call it something else! Yet op model work remains vital, for it is the blueprint for how you operate as an organisation and if you get it right, you will set yourself apart from your competitors. So, this begs the question...
How do you ‘get it right’?
There is no perfect way to design a successful operating model, but here are some guiding rules that should help:
1. Clarify your mission
You have an op model that is there to achieve something. Be clear on what it is. Too many leaders define a model they like without thinking through what it is actually there to achieve. You want a Ferrari? That’s nice – but maybe you actually need a people carrier. Be clear on the mission you are trying to achieve and design for that.
2. Agree constraints for the design upfront
Be clear on what constraints you need to work within, particularly the headcount and cost factors. With too many op models these get agreed at the end, requiring notable last-minute rework that could leave you with a sub optimal answer. This is a surprisingly common issue, with detailed org charts that took weeks and even months to design subsequently getting completely redesigned in the final days.
3. Define your top-level shape first
Focus on the top-level shape of your op model. The next level of detail can be useful, but there is little point in defining it until the top level is established. With the top-level defined, you can also be clearer on who is likely to lead each area and ensure they are then involved and bought into the lower-level design work when the time is right.
4. Involve your future leaders
Leaders like to shape their functions (organisation, department, teams) as they see fit. Give them a fait accompli that is perfect, and they will inevitably want to change it to show their ‘value’. If you know who your likely future leaders are, involve them in the design work so that you minimise this risk of re-work. Equally, if you know who you don’t want on the bus, minimise their input.
5. Expect the design to change
Some organisations get so serious with op model designs that they can spend 6-12 months designing and perfecting the intricate details to get them exactly right. This is nonsense. No plan lasts first contact with the enemy. Apply the 80-20 rule as no matter what you do, the op model design will evolve as soon as you apply it. Build this into your implementation thinking and be honest with everyone that this is what is expected.
6. Ensure people over process
As high performing firms such as Netflix have found, if you want people to perform you need to give them the space to work out the best way to do so. This means being careful not to overly rely on detailed processes that can turn people into mindless drones. High level processes are useful, but detailed processes should only be required for compliance and security type work where fixed actions are needed. And let’s be honest – how many people read and follow the detailed processes anyway? If you want your people to perform, let them think for themselves.
7. Clarify touch points
Op models for larger organisations are all about touch points. Understand and define these, so that it is clear how the different factions engage smoothly with one another.
8. Ask your customers what they want
Often there is the assumption by leaders that they know what their customers want. This is a dangerous assumption that can mean a key need is missed, so always ask your customers what they want. Customers could be a whole range of groups, so be clear on who you need to ask. For example, common customers of your op model could be your actual end customer, other teams, companies, partners or employees. Build their needs into your thinking.
9. If People-related change is high, consider changing the people
For op models that require a notable change to behaviours and mindsets, there is almost always a massive resistance to the change amongst the people involved. This takes time to overcome, and the organisation may not have the luxury of taking the time required. If this is the case, acknowledge the issue and consider strategically changing the people involved. This will mean embedding new hires who can and will champion the changes and removing specific individuals who will inevitably derail you. This is not a popular approach for obvious reasons, but it is realistic and practical one.
10. Engage the People team
It is always tempting to keep the People team at arm’s reach when undergoing op model work. However, they can save you a lot of time by quickly telling you what is and is not possible and flag any other op model work in the organisation that could complement or conflict with your plans.
11. Make the difficult decisions
Op models are sometimes seen as simply moving the deck chairs because they avoid making the difficult decisions. Poor performers are simply moved around, or some roles and teams are kept simply because there is a desire to keep a specific individual. Make the hard decisions. During the Covid-19 pandemic, many organisations were forced to make these decisions at speed and are now reaping the rewards. Are such decisions painful? Yes, but this is part of the job for senior leaders. You make the tough calls.
12. Don’t throw the baby out with the bath water
If it is broken, try fixing it rather than starting from scratch. A frustration many people have with op models is that they throw away something that wasn’t quite working, rather than spending the time to actually get it right. A classic example of this is project delivery frameworks. Over time, inefficiencies build up around these, either due to being over or under-governed. The framework gets blamed and a new one is implemented. The truth is that the framework is normally fine, just poorly managed. Whilst not as much fun, sometimes the best answer is to persevere.
13. Think ahead to implementation
13 is unlucky for some, and so is the implementation roadmap. An op model can be perfect but impossible to implement, making is worthless. The challenges with implementations need to be considered up front or else you risk them killing months of detailed op model work. The implementation approach and roadmap should have influences on your design. For example, removing department X before department Y is ready to take on X’s responsibilities is a recipe for disaster, or trying to implement too many new roles for the People Team to recruit for. Implementation is as important, if not more so, than the op model itself. It should also factor in rule 5 – expect the design to change.
14. Ensure there is a bought into business case
If you want to make notable and meaningful op model changes, make sure there is a business case and that it has the right backing. Failure to get this will mean watered down new models (normally removing all the real benefits) or no new model at all. This should be clear on the mission (rule 1), reflect what customers want (rule 8), note the constraints (rule 2), and ensure there is support for making the difficult decisions (rule 11). The case should also set expectations on the implement approach and implications (rule 13).
Following these rules will increase your chances of success with your op model, although there are no guarantees. What is guaranteed is that whether you design it or not, you will have an op model your organisation is following. Better to follow one you have consciously defined and assessed, rather than one you subconsciously are all following without fully understanding.
