Revenue beat plan. So why did we miss on operating income?
Golden walks your GL from revenue to operating income, colors every variance red or green, and points at the account that broke the half.
The board deck is due and the numbers disagree
Dana leads FP&A at a software company with three business units. H1 FY2025 revenue beat budget, opex looks fine, and operating income still came in almost 10% short.
Leadership needs one answer: re-forecast, or call it a one-off. Finding it by hand means pivoting thousands of GL rows by account, entity and month until something jumps out.
Sign it once, read it everywhere
Costs live in the GL as positive numbers. One calculated field flips them, so a positive variance means favorable for revenue and costs alike, and the waterfall runs cleanly to operating income.
From there Golden gives you the P&L walk, a conditionally formatted variance pivot and a drill to the single account that matters.
How Golden answers it
Walk the P&L
A waterfall from Revenue through COGS, Sales & Marketing, R&D and G&A, ending at operating income. Build a second one on budget and compare the last bars.
Turn the variance pivot red and green
Account group and account down the side, months across, a diverging color scale on Variance %. The Cloud Hosting row turns red in March 2025 and stays red.
Plot the culprit
A line of Cloud Hosting actual against budget. The two lines separate in March and never come back.
Check whether it's everywhere
Variance by entity, colored by account group. Every business unit shows a red COGS bar, so this is a shared hosting cost problem, with no single team to blame.
One account explains the half
- 01H1 revenue: $28.72M against a $28.37M budget, 1.2% favorable.
- 02Operating income: $2.05M against $2.27M, a $216,851 miss (9.6%).
- 03Cloud Hosting ran $701,944 over budget in the half, 21.9% over from March through June. That one account is bigger than the entire miss.
- 04G&A came in 4.1% under, which is part of why the opex block looked healthy.
From answer to re-forecast
Run Explain Change on operating income from February to March and the driver tree lands on Cloud Hosting across all three entities. Forecast hosting to December to size the full-year overrun if the trend holds.
Then generate a data story for the board deck. It picks up the revenue beat and the hosting overrun as its two beats, which is exactly the slide Dana would have written.
Find your variance before the board does
Bring your GL actuals and budget and get to the account behind the miss.