SK EN
Consulting 8 min read

When every decision waits for the owner: a practical operating model for growing SMEs

How to distribute decision authority, establish a useful management rhythm and remove a bottleneck without creating unnecessary bureaucracy.

AI-generated image: leaders at a growing Slovak company handing over responsibility for a decision
Delegation becomes reliable when authority, boundaries and feedback are explicit—not when responsibility is simply pushed down the organisation.

In the early years of a company, the founder's involvement is an advantage. Customers receive quick answers, priorities remain visible and one person can connect details across sales, delivery and finance. Growth changes the equation. More employees and projects create more decisions, while the founder's day still has only the same number of hours.

The resulting bottleneck often looks like commitment: the owner joins every important meeting, approves exceptions, rewrites offers and resolves conflicts between teams. In reality, the organisation is learning to wait. Managers prepare recommendations but hesitate to act. Employees escalate because the safest answer is “ask the boss”. The founder becomes busier while the company becomes slower.

This matters in an economy shaped by small firms. The European Commission's 2025 Slovakia SME fact sheet estimated that SMEs represented 99.9% of enterprises in the non-financial business sector in 2024, provided 74.5% of employment and generated 53.6% of value added. The Slovak Business Agency likewise describes SMEs as a central part of the national economy. Their ability to grow without losing clarity is therefore not a niche management concern.

Recognise a decision bottleneck before changing the organisation chart

Slow decisions are frequently blamed on the wrong structure or the wrong people. Sometimes that diagnosis is correct. Often the roles on paper are reasonable, but the actual decision process is unclear. Look for observable symptoms:

  • routine matters regularly wait for one senior person;
  • the same question is reopened after different meetings;
  • several people can block a decision but nobody can make it;
  • employees receive responsibility for an outcome without authority over the relevant choices;
  • exceptions silently become the normal process;
  • the owner learns about problems only when they are already urgent.

Do not begin by adding layers of management. First examine twenty to thirty decisions from recent weeks: pricing exceptions, supplier changes, hiring, customer complaints, investment requests, scheduling conflicts or product adjustments. Record who initiated each decision, who supplied input, who decided, how long it waited and why it was escalated. Patterns become visible quickly.

Separate three levels of decision

A useful model distinguishes decisions by their consequences rather than by job title.

  1. Strategic decisions change direction, risk appetite, major investment or the company's market position. Owners or the executive team normally retain these.
  2. Cross-functional decisions affect several teams—for example capacity allocation, key-account priorities or a change to the delivery model. They need one named decision-maker and structured input from the affected functions.
  3. Operational decisions apply agreed policy in recurring situations. They should usually be made close to the work, within clear limits.

The purpose is not to classify every possible event. It is to make the common and consequential decisions predictable. If people know that a team lead may resolve a customer issue up to an agreed cost, response time and contractual risk, most cases stop travelling upwards. The unusual case still receives senior attention.

Give every recurring decision five elements

A short “decision card” is often enough. It should state:

  • the decision: what choice is actually being made;
  • the decision owner: one person who has the final call;
  • required input: who must contribute facts or consequences before the choice;
  • boundaries: budget, risk, quality, legal or customer limits;
  • escalation triggers: the conditions that move the matter to a higher level.

One final decision-maker does not mean one-person thinking. Colleagues may challenge assumptions and provide specialist input. The distinction is that consultation has an endpoint. Consensus can be valuable when commitment is more important than speed, but requiring consensus for every operational matter gives everybody a veto and nobody clear accountability.

Delegation needs boundaries and a feedback loop

“Take ownership” is not a delegation system. A manager needs to know the result expected, the choices they may make independently, the resources available and the point at which the risk exceeds their mandate. The senior leader needs a way to see outcomes without pre-approving every step.

Boundaries should be concrete enough to guide action. Instead of “keep discounts reasonable”, specify the margin, contract term or customer category that requires review. Instead of “hire when necessary”, agree the approved role, compensation range and capability criteria. Good boundaries reduce anxiety for both sides: the delegate can act, and the owner knows that defined risks remain visible.

Review the first few decisions shortly after they are made. Ask what information was missing, which boundary was unclear and what can be improved. This is coaching, not retroactive approval. If a reasonable decision produced a disappointing result, changing the rules may be appropriate; punishing the person for acting within the agreed mandate teaches the whole organisation to escalate again.

Create a management rhythm that resolves—not reports

More meetings do not create more control. A small set of meetings with distinct purposes is usually more effective:

  • Weekly operational review: exceptions, immediate constraints and decisions required within days.
  • Monthly business review: performance against a small number of priorities, important risks, capacity and cross-functional trade-offs.
  • Quarterly direction review: assumptions about markets, capabilities, investments and what the company will stop or change.

Send routine information before the meeting. Use meeting time for disagreement, trade-offs and decisions. For each item, record the decision, owner, due date and the assumption that would justify reopening it. A decision log prevents the company from debating the same topic repeatedly without new evidence.

Use data as input, not as a substitute for judgement

Digital tools can make relevant information available faster, but a dashboard does not decide which risk is acceptable or which customer promise the company intends to keep. Slovakia's 2026 Digital Decade report notes progress in SME digitalisation alongside a continuing gap to EU peers and shortages in digital skills. Technology should support a clear operating model, not automate confusion.

For an important choice, a one-page brief can be more useful than a large presentation: the decision required, available options, key evidence, assumptions, risks and recommendation. Mark facts separately from estimates. State what is not known. This improves the quality of discussion and creates a record that can be reviewed later without manufacturing false certainty.

Measure whether the bottleneck is actually shrinking

Revenue and productivity have many causes, so they are poor short-term proof of a decision redesign. Begin with closer operational indicators: the waiting time for common decisions, the share resolved at the intended level, the number reopened without new evidence, repeated escalations and founder interventions in routine work.

Also ask managers and employees two direct questions: “Which decisions are you expected to make?” and “Which decisions do you still avoid because the boundary is unclear?” If answers differ sharply between levels, the formal model has not yet become the real one.

The broader economic evidence supports taking management capability seriously. The OECD's 2026 productivity compendium reports that SME labour productivity averaged 65% of large-firm productivity across OECD and accession countries in 2024, while emphasising that results vary substantially by country and industry. It also identifies ownership and management structures as relevant factors. OECD work focused on Slovakia has previously called for stronger workforce and management skills, digital adoption and the ability of firms to operate in wider markets. None of this proves that a decision card creates productivity. It does explain why professionalising management deserves attention alongside finance and technology.

Avoid replacing dependence with bureaucracy

The failure mode on one side is founder dependence; on the other, a process for everything. A ten-person company does not need the governance machinery of a multinational. Document decisions that are frequent, costly, risky or cross-functional. Leave simple reversible choices close to the work. The test is whether the structure makes a good decision easier, not whether every box in a framework is complete.

External consulting can help by making hidden patterns visible, facilitating difficult agreements and bringing a neutral view to role conflicts. It should not become a permanent substitute for management responsibility. The organisation has to own the model, understand the trade-offs and continue improving it after the workshop ends.

More information about our approach to organisational clarity and practical implementation is available on the Consulting page.

Sources and further reading

Facts and research context were checked against the following sources on 23 August 2026. The operating model, diagnostic questions, decision-card structure, implementation plan and wording in this article are original editorial work by Merkle s. r. o.