Org Design for Small Companies: Andrew Wilkinson & Tiny on Roles, Spans & Layers
Org Design for Small Companies: Andrew Wilkinson & Tiny on Roles, Spans & Layers
Org Design for Small Companies: Andrew Wilkinson & Tiny on Roles, Spans & Layers is my plain-English playbook for building a team that ships fast without burning out.
I’ll show you how I set roles, spans, and layers, when I add managers, and how I keep the org simple as we scale.
I’ll keep every sentence short and on its own line.
I’ll give you copy-paste templates you can use today.
First principles I won’t bend
Structure follows cash flow and customer value, not vibes.
Titles follow scope, not tenure.
Managers exist to increase output, not to attend meetings.
Fewer layers.
Wider spans.
Clear owners.
For writing that keeps everyone aligned, see: I Don’t Respond to Long Emails.
My rule of thumb for spans & layers
Target 5–8 direct reports for steady teams.
Peak at 10 only during short sprints with clear SOPs.
Keep layers to 3 or fewer until ~120 people.
If a decision passes more than 2 managers, I flatten it.
When to add the first manager
Add a manager when one person spends >50% of their week unblocking others.
Or when quality drops because nobody owns code review, QA, or creative edits.
The role is a force multiplier, not a promotion gift.
The “small org” ladder that works
IC → Senior IC → Lead (player-coach) → Manager → Manager of Managers.
IC and Manager tracks are parallel.
Pay the best ICs more than average managers.
Retain creators.
Don’t force management.
Role scorecards beat job posts
Define Mission, Outcomes, Competencies, Interfaces, Scope/Not-Scope.
Three outcomes, not twelve.
Tie each outcome to a metric and a review cadence.
Copy/paste template below.
Team charters stop turf wars
Write Purpose, Customers, Inputs, Outputs, Definition of Done, Interfaces.
One page.
Publish it.
Link it in tickets.
For simple storytelling your team will read, see: Never Tell, Always Storytell.
Meeting architecture in one minute
One weekly team checkpoint (30–45 min).
One cross-functional sync (30 min).
Daily async updates in one channel.
No standing 1:1s unless you manage people.
Use meetings to decide, not to update.
The operating loop that keeps orgs light
Weekly: metrics, decisions, next 7 days.
Monthly: strategy, budget, org risks.
Quarterly: scorecards, promotions, spans & layers check.
For cadence that compounds, read: 02: Journaling With AI.
Headcount budgeting without drama
Budget by outcome, not seat.
Write the business case for each role in 5 lines.
If it doesn’t pay back in ≤12 months (or defend core), wait.
The “Org Debt” checklist
Too many layers for headcount.
Managers with ≤3 reports for months.
Orphaned teams with no P&L owner.
Undefined interfaces causing ping-pong.
Fix org debt like tech debt: small, constant refactors.
Spans & layers by company size (examples)
5 people.
No managers.
Founder runs priorities.
Everyone ships.
15 people.
2 Leads (player-coaches) each with 5–7 reports.
Founder manages Leads.
40 people.
Functional heads (Eng, Product, Rev, Ops, Finance).
Each head spans 6–8.
Two layers total.
80 people.
Add one layer where load is real (e.g., Eng Managers).
Keep total layers ≤3.
Introduce a lightweight Program role only if projects cross 3+ teams.
RACI when decisions slow down
Responsible ships.
Accountable approves.
Consulted gives input.
Informed stays out of the way.
If more than one A, you have no A.
Copy/paste — Role scorecard (one page)
ROLE: {Title} TEAM: {Function} MANAGER: {Name} VERSION: v1.0 — {YYYY-MM-DD}
MISSION
In one sentence, why this role exists.
TOP 3 OUTCOMES (with metrics)
1) {Outcome} — Target: {number/time/quality}
2) {Outcome} — Target: {…}
3) {Outcome} — Target: {…}
COMPETENCIES (5–7 bullets)
• {Skill/Domain}
• {Tool/SOP}
• {Behavioral norm}
INTERFACES
• Works with {Teams/Roles} on {Topics}. Decision owner: {Name}.
SCOPE / NOT-SCOPE
• Owns {A, B, C}.
• Does not own {X, Y, Z}.
CADENCE
• Weekly 30-min checkpoint with {Manager}.
• Monthly metrics review.
• Quarterly scorecard refresh.
LEVELING
IC ↔ Lead ↔ Manager tracks documented. Promotion requires evidence on outcomes + scope.
Copy/paste — Team charter (one page)
TEAM: {Name} LEAD: {Name} VERSION: v1.0 — {YYYY-MM-DD}
PURPOSE
What we do and for whom.
CUSTOMERS
Internal: {Teams}. External: {Segments}.
INPUTS → OUTPUTS
We take {inputs} and produce {outputs} with {SLAs}.
DEFINITION OF DONE
What “done” means for our deliverables.
METRICS
North Star: {metric}. Guardrails: {quality/time/cost}.
INTERFACES
We rely on {Team A} for {X}. We provide {Y} to {Team B} by {cadence}.
ESCALATION
Who decides when we’re stuck and in how long (T+24h).
CHANGE LOG
{Date} — updated scope/SLAs.
Copy/paste — Span & layer review (quarterly)
DATE: {YYYY-MM-DD}
HEADCOUNT: {#}
SPAN SNAPSHOT
• Managers with <4 reports: {#} → Fix by {date}.
• Managers with >10 reports: {#} → Add lead or SOP and recheck.
LAYERS
• Total layers from CEO to IC: {#} (Target ≤3).
• Decisions needing >2 approvals: {#} → Flatten or delegate.
RISKS
• {Team} single point of failure → Cross-train {Names}.
• {Interface} ping-pong → Add RACI and SLA.
ACTIONS (30 days)
1) {Action} — Owner {Name} — Due {date}
2) {Action} — Owner {Name} — Due {date}
Copy/paste — Reorg announcement (calm, human)
Subject: Simple org updates so we move faster
We’re simplifying the org to ship more with less friction.
What’s changing.
{Team A} now reports to {Lead}.
{Team B} merges into {Team C}.
No title changes beyond {list}.
Why.
Fewer layers, clearer owners, faster decisions.
What’s not changing.
Roadmap, compensation, and goals.
If you’re impacted, your manager will meet you today with details.
Thank you for staying focused on customers while we tidy our house.
Promotion and scope increases
Promote for scope and outcomes, not time served.
Write a promotion packet: before/after scope, metrics evidence, interfaces added, risks now owned.
No “title inflation” as a retention hack.
Manager load sanity check
New managers cap at 6 directs for 90 days.
Player-coaches cap at 5.
If a manager’s calendar is >60% meetings, something’s broken.
Fix with delegation, SOPs, or IC promotions.
How I staff cross-functional work
I assign a DRI from one home team.
I borrow ICs with their home managers’ blessing.
I run the project with a one-page brief, weekly demo, and a clear kill switch.
Programs do not become permanent without proof.
Remote, hybrid, in-person rules
Remote requires asynchronous clarity and stronger SOPs.
Hybrid requires office days with a purpose.
In-person requires short docs or you’ll rely on memory.
Pick one model and design for it.
Don’t blend all three.
Org metrics I watch monthly
% managers of total headcount.
Median span of control.
Layers CEO→IC.
Time from decision request → decision.
% goals green at quarter end.
If these trend well, the org is healthy.
Red flags I fix fast
Two managers “sharing” a team.
Committees with no DRI.
Roadmaps owned by nobody.
Managers without 1:1s or metrics.
Career paths that only go up, never deep.
Green flags that accelerate everything
Player-coach leads who ship.
ICs paid like stars for hard impact.
Scorecards that fit on one page.
One weekly operating loop everyone respects.
Simple org chart you can explain in 60 seconds.
Example tiny org charts you can steal
Product-led SaaS (15 people).
CEO → Product Lead (PM, Design), Eng Lead (6 ICs), Rev Lead (Sales 2, CS 2), Ops/Finance (1).
Brand/e-commerce (15 people).
CEO → Growth Lead (Paid, Email/SMS, Site), Brand Lead (Creative, Merch), Ops Lead (3PL, Supply, CX), Finance (1).
Marketplace (40 people).
CEO → Supply Lead (Success, Vetting), Demand Lead (Growth, CRM), Product Lead (PM, Design), Eng Director (EMs + ICs), Ops/Trust Lead, Finance.
Scripts for org hygiene
Span fix DM.
“Let’s move {Name} under {Lead} this quarter.
Target span 7–8.
No title change.
I’ll announce Friday.”
Scope swap email.
“{Team} will now own {Outcome}.
{FormerTeam} remains consulted.
DoD and SLAs attached.
Goes live Monday.”
Manager reset.
“Your job is output and growth of your people.
Ship weekly, review monthly, promote on scope.
If a meeting doesn’t move work, cancel it.”
FAQs
What’s the best span of control for small companies.
Aim for 5–8.
Use player-coaches before layering managers early.
When should I add a new layer.
When a manager has >10 stable directs and quality drops, or when coordination overhead eats the week.
How do I keep ICs from leaving for titles.
Create a deep IC ladder with pay parity and prestige.
Celebrate impact, not headcount.
How do I know it’s time to reorg.
Decisions stall, quality drops, and owners are fuzzy.
Run the span & layer review and act in 30 days.
How many meetings should managers have.
As few as needed to decide and coach.
Cap recurring meetings and move updates to async.
Where do promotions go wrong.
Promoting for loyalty or time served.
Promote for scope + outcomes backed by evidence.
How do I onboard the first manager.
Give a scorecard, clear span, and a 30-60-90 plan.
Shadow their 1:1s in month one.
What do I do with a brilliant IC who hates meetings.
Protect their maker time.
Pay them top of band.
Make senior IC a first-class path.
Can I flatten later without chaos.
Yes.
Move reporting lines, keep scopes stable, announce calmly, and review in 30 days.
What kills orgs at 40–80 people.
Too many layers, title inflation, and cross-team ping-pong.
Fix with charters, DRIs, and one operating loop.
Conclusion
Org Design for Small Companies: Andrew Wilkinson & Tiny on Roles, Spans & Layers is about keeping layers low, spans sane, and ownership obvious so the work ships and the team stays calm.
Use the scorecard, the team charter, and the span review each quarter, and you’ll scale with fewer meetings and more output.
Get Your Copy of Never Enough at https://www.neverenough.com/
