A guide from a company that keeps resetting — NZ$980m, at 89%.

We lock a binary: Spark New Zealand's reported adjusted EBITDAI for FY2027 is NZ$980m or higher. Confidence 89%.

Guidance is NZ$1,010-1,080m. Our threshold sits 3% below the floor of that range and 5.3% below what FY2026 actually delivered.

The method across this series leans on company guidance because management has information we do not and a reputational cost for missing. That reasoning weakens when a company has recently and repeatedly failed to hold its own expectations, and Spark is in that position. FY2026 adjusted EBITDAI fell 2.4%, the company itself characterised the period as one of the most challenging in its history, and it has run successive rounds of job reductions while exiting its data centre business.

The forward guide is also structurally awkward. The NZ$1,010-1,080m range straddles the NZ$1,035m just delivered, which means management is guiding to roughly flat performance while simultaneously absorbing the loss of data centre earnings and continued digital services margin pressure. Getting to the midpoint requires the mobile growth and productivity benefits to more than offset two known drags. That is a coherent plan, not a fantasy — but it is a plan, and this company's recent record of converting plans into results is what the 89% is pricing.

There is a genuine floor under it. Free cash flow rose 18.5% to NZ$308m and the dividend was set at 100% of free cash flow, which imposes discipline. Mobile service revenue is growing again. New Zealand telecommunications is a stable three-player market with inertial revenue. A 5.3% decline from FY2026 would be a materially worse year than the one management is planning for.

The metric is adjusted EBITDAI, with the I for the impact of certain items Spark defines in its reporting. It is not EBITDA, and it is not statutory earnings. The criterion names it.