Service pillar
From a 50-person business to a 300-person machine.
Operating model, process, org design and the management layer that has to exist before the headcount does.
In short
Operations scale-up advisory at Keel Partners rebuilds how work flows through a business that has outgrown its founding operating model. We redesign the operating model, document the processes that matter, build the missing management layer, and put the data and tooling underneath it — so delivery quality, cash and culture survive tripling the headcount.
Fit
Who this is for
The 50-to-300 transition breaks businesses that were working perfectly well at 50. It is a structural problem, not an effort problem.
Right fit
- Services, healthcare, retail and light-manufacturing businesses between roughly 40 and 300 people
- Companies whose delivery quality has started slipping as volume rises
- Businesses where the founders are still the escalation path for operational decisions
- Multi-site operations where each site runs a different version of the process
- Groups that have just acquired and now run two incompatible operating models
Not a fit
- Businesses that need headcount reduction rather than scale
- Pure software engineering organisations — that is not our operating experience
- Anyone wanting an ERP selection and implementation as the engagement
- Situations where the real problem is demand, not delivery
Problems
The problems we are usually called on
Quality slips with volume
The same team, the same process, twice the throughput — and complaint rates rise faster than revenue.
No middle management
Twelve direct reports to a founder, and nobody in between who can hire, coach or say no.
Every site does it differently
Three branches, three processes, three definitions of the same metric.
Cash tightens as revenue grows
Working capital lengthens invisibly: stock, receivables, work-in-progress nobody owns.
Tribal knowledge
The process lives in the head of one person who has been there since the start.
Tooling bolted on
Six systems, none integrated, and a spreadsheet that is the real source of truth.
Method
How we work
Ten to sixteen weeks, run in the business rather than in a workshop room. We spend the first fortnight on the floor.
Map the flow
Follow the actual work from order to delivered to paid. Where it queues, where it loops back, where it depends on one person.
Find the real constraint
Capacity, quality, cash or decision speed — usually one dominates and the rest are symptoms.
Redesign the operating model
Roles, hand-offs, decision rights and service levels. Written down at the level someone new could follow.
Build the management layer
The supervisory tier that has to exist before headcount can grow: scorecards, one-to-ones, escalation rules, hiring bar.
Instrument it
The five operating metrics, where the data comes from, who reads it and when. Analytics and tooling applied to the process — not a systems project.
Run it and adjust
We stay through at least one full cycle. Designs that have never met a bad week are not designs.
Deliverables
What you get
Artefacts a team can run without us in the room.
- A documented operating model: flow, roles, decision rights, service levels
- Standard operating procedures for the processes that actually decide quality
- An org design for the next stage, with a hiring sequence and scorecards
- An operating dashboard with five metrics, defined sources and named owners
- A working-capital view: where cash sits in the cycle and who owns each element
- A tooling recommendation that fits the process rather than replacing it
Geography
UAE, KSA and Europe are not the same problem
Scaling in the Gulf has constraints European operators do not expect, and vice versa.
| United Arab Emirates | Saudi Arabia | Europe / UK | |
|---|---|---|---|
| Labour model | Visa-sponsored workforce; hiring speed is high, but every change carries a sponsorship and gratuity consequence | Saudisation quotas shape the org design itself | Notice periods and consultation obligations slow every restructure |
| Management bench | Thin; the supervisory layer usually has to be built rather than hired | Thin and quota-constrained | Available but expensive and slow to move |
| Multi-site | Emirate-level licensing and inspection differences between sites | Regional expansion carries separate registration steps | Regulatory variation across countries dominates |
| Cash effect of growth | Gratuity accrual grows with headcount and tenure and is routinely under-provisioned | Similar end-of-service obligations | Pension and social contributions are predictable but heavy |
Selected work
Mandate shapes
The shape of a scale-up mandate, and the operating background behind it.
Multi-site services operation — typical mandate
Fourteen weeks: flow mapping across sites, a single operating model, SOPs for the quality-critical processes, a supervisory layer with scorecards, and a five-metric dashboard. One full operating cycle of partner-chaired review afterwards.
Typical mandateOperating background
The partner group carries over 100 years of cumulative industry experience across retail, FMCG, healthcare, hospitality and niche manufacturing, including multi-store retail and diagnostics operations run first-hand — plus 23 years of enterprise technology and process work.
Why usFAQ
Questions buyers actually ask
Is this a lean or six-sigma programme?
No. We borrow from those toolkits where they help, but a 60-person business does not need a belt programme; it needs decision rights, a supervisory layer and five metrics somebody reads on a Monday.
Do you implement an ERP?
No. We will tell you what the process needs and what a system would have to do to support it, and we will sit with you while you choose. Selecting and implementing the system is a specialist job and it is not ours.
How long does it take?
Ten to sixteen weeks for the redesign, then one full operating cycle of review. Multi-site adds time roughly in proportion to how differently each site currently works.
What is the most common root cause you find?
Decision rights. Not process documentation, not systems — the absence of a written answer to who can decide what, and at what value. Everything queues at the founder.
Will this cost us people?
This pillar is about growing without breaking, not cutting. Roles change and some people do not want the new version of their job, which is worth planning for honestly rather than discovering in month three.
How do you handle culture?
By being specific rather than aspirational. Culture at this stage is mostly the sum of what gets escalated, what gets tolerated and who gets promoted. We work on those three things and leave the poster on the wall alone.
Can you work across sites in different emirates or countries?
Yes, and multi-site is where most of the value sits. The first deliverable is usually a single definition of the process and the metric, which frequently does not exist.
What if the real problem is that we do not have enough demand?
Then this is the wrong pillar and we will say so in the first fortnight. Go-to-market is a different engagement, and building capacity for demand that is not there is an expensive way to find out.
Do you bring a team?
A partner plus, where the mandate needs it, one or two named associates. If we cannot cover the file with a partner every week, we do not take it.
What does it cost?
A fixed fee against a defined scope, quoted after a short diagnostic. We will tell you which parts of the work your own team can do — that usually reduces the fee, and it always improves adoption.
Next step
Book a 30-minute briefing
Bring the decision you are stuck on. We will tell you what we would do and whether we are the right firm to do it.