Bring this list to your first 2026 meeting
Your holiday homework...
January board meetings have a habit of going one of two ways.
Either they stay safely high level. Growth targets. Channel performance. A slide on “AI opportunities”. Everyone nods. Nothing really changes.
Or they get dragged into the weeds. One metric doesn’t reconcile. Someone questions stock. Returns get mentioned, briefly, then parked. The meeting overruns and the actual decisions get deferred.
Neither is especially useful.
If 2026 is going to be different, the first board meeting needs a different shape. Not more slides. Not more data. A better list.
What follows is the list we would bring, based on the themes that have come up again and again across Commerce Thinking articles over the last year. ERP performance under pressure. Returns as an economic system. Inventory truth. Automation reality. Ownership. Trust.
Bring this list. Use it to structure the meeting. And be honest about the answers.
The list
1. Which numbers do we not fully trust, and why?
This sounds uncomfortable. It is meant to be.
Every commerce business has at least one KPI that gets debated instead of used. Revenue by channel. Stock on hand. Return rate. Contribution margin. Forecast accuracy.
The question is not whether the number is right.
The question is why people doubt it.
Is it timing lag?
Multiple systems?
Manual adjustments?
A workaround everyone relies on but no one owns?
We have written extensively about this in the context of ERP implementations and why “a single source of truth” is rarely the real issue. The real issue is whether people believe the data enough to act on it.
Bring to the meeting:
A short list of disputed metrics
Where they come from
Who adjusts them
What decisions are delayed because of that distrust
If the board cannot agree on what is true, they cannot agree on what to do next.
2. What breaks under peak, and what does that cost us per hour?
Steady state performance is a comfort blanket.
Peak is where the truth comes out.
We have seen this repeatedly in work around ERP lag, order backlogs, warehouse de-syncs, and customer service overload. Systems that look fine most of the year quietly fall apart when volume spikes.
Ask very directly:
What failed or nearly failed during the last peak?
How long did it take to detect?
How long did it take to fix?
What was the financial and CX cost per hour?
This is why peak performance has been such a recurring theme in Commerce Thinking articles. Because it exposes operational debt that does not show up in monthly reporting.
Bring facts, not anecdotes.
This is where investment cases become obvious.
3. What is the true cost of returns, fully loaded?
Returns should never be a footnote.
They touch margin, cash flow, warehouse capacity, customer behaviour, sustainability, and brand trust. Treating them as a CX policy instead of an economic system is one of the most expensive mistakes we see.
At this meeting, you want answers to:
Cost per return, end to end
How much is recoverable and how much is written off
What percentage is habitual or avoidable
How current policy influences customer behaviour
This connects directly back to articles on paid returns, returns strategy, and the true cost of returns. The point is not whether to charge or not. The point is whether the board understands the trade-offs.
If returns cost more to process than the margin on the original order, that is not a CX discussion. It is a commercial one.
4. Which inventory becomes worthless on a date?
Not slow moving.
Not aged.
Worthless.
End of season.
End of contract.
Regulatory changes.
Brand changes.
Market exits.
This came up repeatedly in discussions around merchandising, selling periods, and critical path planning. Especially in categories where timing is unforgiving.
The board should be able to answer:
Which SKUs have a hard sell-by date?
What is the liquidation plan before we reach it?
Who owns the decision to accelerate, discount, or kill?
If the answer is “we’ll deal with it later”, you already know how that ends.
5. Who actually owns execution, not just strategy?
This is where many board discussions quietly fail.
Everyone agrees on the direction. No one owns the outcome.
We have written about this in the context of ERP projects, vendor-led roadmaps, and blurred accountability. Strategy without clear ownership turns into a dependency on external partners to make decisions for you.
Bring clarity on:
Who owns delivery timelines
Who defines “ready”
Where vendors advise versus decide
Where internal teams defer because ownership is unclear
Boards do not need more strategy decks.
They need fewer ambiguous owners.
6. What are we calling AI that is actually just automation?
This deserves its own slot, because it keeps derailing serious conversations.
Most of what is being labelled AI in commerce today is workflow automation, rules engines, or long-standing statistical modelling. Useful, yes. Magical, no.
The real shift has been cost and accessibility, not intelligence.
Ask:
Which manual tasks happen 100 times a week and still exist?
Which processes happen at volumes that low-code tools should never touch?
Where is “AI” language masking a lack of process design?
This ties directly back to recent discussions on automation, low-code tooling, and why understanding execution frequency matters more than chasing buzzwords.
The board does not need an AI strategy.
It needs a decision framework for when automation is worth doing.
How to run the meeting
Use the list to structure the agenda.
Five minutes to frame the risks.
Fifteen minutes on the biggest margin leaks.
Fifteen minutes on execution constraints.
Ten minutes agreeing three bets for 2026.
Five minutes agreeing one thing you stop doing.
Not ten bets. Three.
And one deliberate stop.
What good looks like by the end of Q1 2026
If this meeting does its job, you should leave with:
Fewer metrics, trusted more deeply
A clear position on returns economics
One peak fragility removed, not discussed
Clear ownership of delivery, not shared responsibility
Automation decisions based on volume and value, not hype
None of this is revolutionary.
That is the point.
It is the accumulation of lessons that most commerce teams learn the hard way. The advantage is bringing them into the boardroom early, while they are still decisions rather than post-mortems.
Bring the list.
Have the uncomfortable conversations.
And make 2026 a year of fewer surprises.
Really solid framework here. The question about which numbers people don't trust is a game changer for most boards tbh. I've sat through too many meetings where eveyrone's nodding at dashboards but afterwards they're all cross-checking in Excel. What stands out is the distinction between low-code automation and actual process volumes that need proper engineering. At my last company we tried to patch high-frequency order reconciliation with Zapier and it blew up within two weeks during a promotion. Sometimes the "boring" answer is just build it right.