How Do You Forecast H2 Demand After the June Holiday Peak? (Singapore SME Guide 2026)

To forecast H2 demand after the June holiday peak, strip the one-off spike out of your numbers first, then rebuild your forecast on a normalised weekly run-rate drawn from your POS and sales history — not on last month's inflated totals. The June school holidays and the Great Singapore Sale pull demand forward and upward, so treating June as your new normal will leave you over-ordered, over-staffed, and short on cash by Q3. The reliable approach is to separate seasonal demand from underlying growth, anchor your baseline to comparable non-peak periods, and adjust for what you actually know about H2.

Why can't you just extrapolate June's numbers into H2?

Because June is a structural outlier for most Singapore SMEs. The mid-year school holidays drive footfall in F&B, retail, enrichment and leisure, while the Great Singapore Sale compresses discretionary spend into a few weeks. Demand that would otherwise have landed in July or August gets pulled forward, so a naive month-on-month projection double-counts it.

There are three distortions hiding in a June peak. First, pull-forward: customers buy now because of promotions, leaving a trough afterwards. Second, price distortion: discounted unit prices inflate volume but not margin, so revenue and units tell different stories. Third, capacity ceilings: if you sold out or hit a service limit, your real demand was higher than recorded — a problem if you forecast off sales rather than demand. Extrapolating the headline number ignores all three.

What data do you actually need to forecast H2 demand?

You need less than most owners assume, but it has to be clean. The minimum useful set is twelve months of weekly sales by product or category, your promotion calendar, and a simple record of any stock-outs or fully-booked periods. With that you can compare like-for-like.

• POS or sales exports at weekly granularity — monthly is too coarse to see the peak's shape.
• Last year's H2 (Jul–Dec 2025) as a seasonal template, adjusted for this year's growth rate.
• Your 2026 promotion and event calendar — National Day, school terms, year-end and any planned campaigns.
• Lead times from suppliers, so the forecast translates into order dates rather than just numbers.

If this data lives in three different spreadsheets, a WhatsApp order thread, and a POS dashboard, consolidate it into one table first. A forecast built on scattered sources will quietly inherit every gap in them.

How do you normalise demand after the June peak?

Normalising means establishing what your demand looks like without the spike, then layering known H2 events back on. Work through it in five steps:

1. Set a clean baseline. Take the weekly average from a non-peak stretch — typically February to early May 2026 — and treat that as your underlying run-rate.
2. Measure the peak's lift. Compare your June weeks against that baseline to see how much was genuine extra demand versus pulled-forward sales. A useful tell: if the weeks immediately after the peak dip below baseline, that gap is your pull-forward.
3. Calculate a seasonal index. For each H2 month last year, work out how it compared to that year's baseline. A July index of 0.9 means July typically runs at 90% of normal — common after a June pull-forward.
4. Apply your growth rate. Compare H1 2026 against H1 2025 to find your real year-on-year growth, and scale the seasonal template by that figure rather than by the peak.
5. Overlay known events. Add explicit adjustments for confirmed H2 campaigns, new products, or lost/won contracts. These are judgement calls, but documenting them makes the forecast reviewable later.

The output is a normalised weekly demand figure for July through December — grounded in your baseline, shaped by last year's seasonality, and scaled by this year's growth.

How do you turn the forecast into stock and staffing decisions?

A forecast only earns its keep when it changes a decision. Convert your H2 demand numbers into three operational outputs. For stock, work backwards from supplier lead times so reorder dates land before your projected demand, and hold safety stock only against your high-variability lines. For cashflow, map the expected post-peak Q3 dip so you do not over-commit on inventory or hiring just as revenue softens — this is where forecasting connects to your H2 cash plan. For staffing, match roster hours to the weekly run-rate rather than the June high-water mark, which is often where lean teams quietly overspend.

Set a monthly checkpoint to compare actuals against the forecast. A forecast that is never reviewed is just a guess with a spreadsheet; a forecast you correct each month becomes steadily more accurate and starts to feed your dashboards and any AI workflows downstream.

What tools should a lean Singapore SME use?

Start with what you have. A single, well-structured spreadsheet with weekly columns and a seasonal-index row will out-forecast most expensive software a small team never fully adopts. Once your data is consolidated and the method is proven, graduate to a lightweight BI tool — Looker Studio, Power BI, or your POS provider's analytics — to automate the weekly refresh. The sequence matters: get the single source of truth and the method right first, then automate. Buying a forecasting tool before your data is clean simply produces wrong answers faster.

Frequently Asked Questions

1. How much sales history do I need before I can forecast H2 demand?
Twelve months of weekly data is ideal because it captures a full seasonal cycle, including last year's June peak and H2. If you only have six months, you can still set a baseline and project growth, but treat your seasonal adjustments as rougher and review them more often.

2. Should I forecast in units or in dollars?
Both, separately. Forecast units to drive stock and staffing decisions, and dollars to drive cashflow. Because the June peak is usually discount-driven, units and revenue diverge — relying on revenue alone will over-state the demand you need to resource for.

3. How often should I update the forecast?
Monthly is the practical cadence for most Singapore SMEs. Compare actuals to forecast at month-end, adjust the next months for any variance, and lock in supplier orders against the updated numbers. Reforecasting weekly is usually overkill unless you are in a fast-moving or highly seasonal line.