The 2027 Headcount Plan You Wrote Is Probably Wrong. Here's How To Check.
We wrote earlier this week about why mid-2026 has stopped behaving like a normal slowdown. Short version: unemployment is the highest it's been since 2015 and the Official Cash Rate went up anyway, because this is a cost shock rather than a demand problem, and cutting into it wouldn't help.
That's the diagnosis. This is the part you can actually do something with.
Below is the worksheet we'd run a 2027 headcount plan through. It won't tell you what your plan should say. It will tell you which parts of it are resting on an assumption that has already stopped being true.
Work through it in order. The first three questions do most of the work.
Part one: what your plan is quietly assuming
1. Write down, in one sentence, what your plan assumes about conditions next year.
Not what you'd say in a board paper. What you actually believe. Most plans written in the first half of this year contain some version of "it gets easier from here", and almost none of them say so out loud. If you can't find the assumption, that's the finding. An unstated assumption can't be tested and won't get revisited when it breaks.
2. Ask what specifically has to happen for that assumption to hold.
Rates falling? A particular contract landing? A budget cycle resetting? Name the trigger.
If the answer is anything like "the market picks up", push harder until you have something you could put a date against. Vague triggers never fire, so the plan never gets revised, and you find out in March.
3. Now check whether that trigger is still plausible.
This is where most 2027 plans come apart. If the trigger was cheaper money, be aware that the Reserve Bank raised the OCR to 2.50 percent in July, its first increase in three years, and kept a tightening bias. Bank economists writing in July were forecasting the rate going higher through into 2027, not lower. That's a forecast rather than a certainty, but it's the opposite direction to the one most plans assume.
If your trigger has become less likely rather than more, you don't necessarily need a new plan. You need a second one.
Part two: what the numbers do to your assumptions
4. Check whether your salary bands are actually flat, or only nominally flat.
Wage growth ran at 2.0 percent over the year to June. Inflation ran at 4.1. Holding a band flat in that gap isn't holding steady, it's a real-terms reduction for whoever is sitting in the role.
That may still be the right commercial call. Just make it knowingly, and budget for the retention conversation that follows it rather than being surprised by one.
All figures here are gross. General information only, not tax advice. Talk to your accountant about your own situation.
5. Estimate your screening load, not just your role count.
Applications per job ad have been rising, and underutilisation hit 13.8 percent in the June quarter, which is 440,000 people who are unemployed, want more hours, or are available but not currently looking. That's a much bigger pool than the headline unemployment rate suggests, and a good chunk of it will apply to your role.
Candidate volume isn't your constraint next year. The hours required to sort through that volume might be. If your plan budgets for roles but not for the time to fill them, it's incomplete.
6. Separate the roles you need from the roles you'd like.
Standard advice, but there's a specific version of it this year. A net 10 percent of firms told NZIER they had cut staff in the June quarter, in the same survey where confidence improved sharply. Sentiment and behaviour have come apart.
If you're benchmarking your plan against how optimistic your peers sound, you're benchmarking against the wrong signal. Benchmark against what they're doing.
Part three: timing and what breaks it
7. Mark the fixed points before you plan around them.
The Reserve Bank reports on 2 September, 28 October and 9 December. The next labour market read lands on 4 November. The general election is Saturday 7 November, with advance voting opening on 26 October. And the country largely stops for the second half of December.
Look at what that does to a fortnight. The Reserve Bank's October review, the Q3 labour market figures and the election all land between 28 October and 7 November, with the campaign running through all of it. That's a stretch where a lot of businesses quietly stop making decisions, and it sits directly between you and the December shutdown.
If you need someone in a seat before Christmas, the practical deadline is earlier than the calendar suggests. We've written separately about how much runway those dates actually leave you and why six to eight weeks from first conversation to first day is realistic rather than pessimistic.
8. Ask what you'd do if the next two rate decisions both go up.
Not a forecast. A stress test. If two more increases would force you to cut a role from the plan, better to know which role now, while you can sequence it, than in February when you're choosing under pressure.
Write the name down. That's the whole exercise.
9. Ask what you'd do if you're wrong in the other direction.
Plans built defensively fail differently. If conditions ease faster than expected and your competitors move first, what's your lead time to respond? If the answer is longer than theirs, your defensive plan has its own cost.
How to use this
Nine questions, and you'll get through most of them in an afternoon. The output isn't a new plan. It's a short list of the places your existing one is load-bearing on something that has already changed.
Our experience is that plans rarely fail because someone got a forecast wrong. They fail because nobody wrote the forecast down, so nobody noticed when it stopped being true.
If you'd like a second pair of eyes on a 2027 plan, or you want to know what roles like yours are actually taking to fill right now across Auckland and Wellington, get in touch with the Find team.