Start with the work, not the growth target.

Leadership teams often begin with a question like this:

How do we grow revenue by 20%?

It is a reasonable business question. It is also too abstract to run day to day.

Nobody performs a “20% growth” task. People answer customers, create proposals, review contracts, and ship features. The work is smaller and more concrete:

• Answer customer questions
• Create proposals
• Follow up on leads
• Review contracts
• Build features
• Resolve support tickets
• Generate reports
• Onboard customers

Growth is what you get when thousands of those units of work go well.

That means the better operating question is not:

How do we grow by 20%?

Ask instead:

Where does work happen in this company, and where is effort being wasted, delayed, duplicated, or constrained?

That question gives a team somewhere to start.

Growth Is Not a Task

Revenue growth is an outcome. It comes from local improvements: faster lead qualification, cleaner handoffs, shorter approval paths, better support triage, fewer rework loops, clearer product decisions, and less waiting between teams.

Most companies manage growth through dashboards. Revenue, pipeline, activation, churn, margin, and headcount all matter. But they are lagging indicators. They show what the system produced, not where it lost time and effort.

To change the system, you have to inspect the work.

Look at the smallest repeatable piece of work that creates value or moves it forward. Ask:

• How often does this happen?
• How long does it take?
• How much waiting is involved?
• How many people touch it?
• How often is it redone?
• Which decisions slow it down?
• Which parts require judgment, and which parts are mechanical?

This is where useful improvements begin.

The Unit-of-Work View

This view makes opportunities easier to spot.

No row in this table sounds like a board-level growth strategy. That is exactly why it is useful.

Companies rarely improve by 20% because an executive invents a perfect 20% plan. They improve when teams find and remove many small constraints. Sales saves eight minutes per lead. Support halves triage time. Engineering reduces pull-request wait time. Finance collects invoices faster.

Each change may be small in isolation. Across important workflows, the effect compounds. Work moves faster, customers wait less, and teams spend more time creating value instead of moving work through the system.

That is how operational improvement becomes growth.

AI Changes the Resolution

Before AI, companies mostly optimized processes.

They mapped departments, designed workflows, bought systems of record, defined approval paths, and standardized procedures. That work still matters. But it happens at a broad level: sales, support, hiring, or finance.

AI lets leaders work at a finer level.

Instead of asking, “How do we automate this department?” you can ask:

What are the 500 tasks people perform every day, and which of them can be completed faster, better, or not at all?

Departments are too large to improve with precision. Processes are better, but often still too broad. A unit of work is small enough to change:

• Summarize this customer call
• Classify this support ticket
• Draft this follow-up email
• Extract contract terms
• Compare this invoice to the purchase order
• Generate the first version of a test plan
• Review this pull request for a known class of issues
• Turn meeting notes into decisions and owners

Some tasks can be automated. Others need assistance, redesign, or removal. Some should remain human because they require judgment, empathy, accountability, or context the system lacks.

The value lies in making that choice deliberately.

Do Not Start With “Automate a Department”

“Automate support” is a slogan, not a plan.

The useful version is:

• Which support tickets are repetitive?
• Which require account-specific context?
• Which require policy judgment?
• Which require engineering investigation?
• Which can be resolved with better product UX?
• Which should never have become tickets in the first place?

Now there is a map of the work.

The same applies everywhere. “Use AI in sales” is vague. “Reduce inbound-lead research from 20 minutes to five while preserving qualification quality” is specific. “Improve finance productivity” is vague. “Extract invoice fields and flag mismatches for review” is specific.

Specific units of work create measurable experiments. Vague transformation programs create activity without a clear result.

How to Run a Unit-of-Work Audit

Start with one function. Do not try to map the whole company at once.

Pick a team with high work volume, visible delays, and meaningful outcomes. Then run this audit:

1. List the 20-50 recurring tasks the team performs every week.
2. Estimate frequency, average time, wait time, error rate, and handoffs for each task.
3. Mark each task as judgment-heavy, context-heavy, repetitive, compliance-sensitive, or customer-sensitive.
4. Identify the top five tasks by total effort or delay.
5. Run one small improvement experiment per task.

The experiment does not need to be a full automation project. It can be a template, a better intake form, a decision rule, an AI-assisted draft, a routing change, a checklist, or one less step.

The goal is not to replace the team. It is to help the team complete more useful work with the same effort.

Measure the result in operational terms:

• Minutes saved per unit
• Fewer handoffs
• Lower rework rate
• Shorter cycle time
• Faster first response
• Higher quality at review
• Fewer decisions waiting on one person

Add up the gains over time.

The Leadership Shift

Top-down goals still matter. A company needs direction. Growth targets alone do not tell teams where to act.

Leadership must translate the target into operational questions:

• Where does work wait?
• Where is effort duplicated?
• Where are decisions delayed?
• Where do people repeatedly search for the same context?
• Where do customers experience friction?
• Which tasks happen so often that a small improvement would compound?

This matters even more with AI. It is tempting to chase a big automation story. The stronger advantage often comes from steadily improving the small pieces of work that happen every day.

Growth becomes manageable when it is treated as a portfolio of bottlenecks removed from daily work.