6 months of GMV Max: 617 million VND in spend generated 17.5 billion VND in revenue and 101,940 orders (ROI 28.39)

Context & challenge

The store already had a very high ROI, but ad spend was still small, so total orders and revenue from ads were limited. The question for this period: how much can budget be raised while still keeping efficiency? The brand name is withheld under a confidentiality agreement.

This is the “problem of being profitable”. A high ROI makes owners reluctant to touch anything. But a high ROI on a small budget means total profit is small too.

Goals for 24/12/2024 – 22/06/2025:

• Sharply raise budget to expand order volume.
• Accept a lower ROI, as long as it remains comfortably profitable.
• Maximise total profit, rather than protecting the highest ROI at any cost.

Diagnosis from the data

GMV Max dashboard for 24/12/2024 – 22/06/2025, compared with the previous 6 months:

• Ad spend: 617,623,646 VND (+1,283.83%).
• Orders: 101,940 (+382.6%).
• Cost per order: 6,059 VND (+186.75%).
• Revenue: 17,533,121,747 VND (+353.75%).
• ROI: 28.39 (−67.21%).

The previous 6 months worked out backwards (estimates, calculated from the dashboard data):

The table shows a pattern businesses should know before raising budget:

• Spend rose 13.8 times, revenue rose 4.5 times. ROI fell from a very high level (about 86 in the previous 6 months, derived from the 67.21% drop) to 28.39.
• In return, scale grew sharply: from about 21 thousand orders to 101,940, and revenue from about 3.9 billion VND to 17.5 billion VND.
• For most categories, ROI 28 is still a healthy profit — so the right goal at this stage is to maximise total profit, not to hold the highest ROI.

The bottleneck: cost per order rose 186.75%. In the previous period, the algorithm only needed to serve the easiest-to-buy customers. With a bigger budget, it had to find more buyers among groups that are harder to convert.

Growth strategy

This is the framework LDH Media applies when raising GMV Max budget for a store that already has a high ROI.

Phase 1 – Foundation: know your limits

• Calculate break-even from each product’s margin to know the minimum acceptable ROI.
• Check inventory and delivery capacity before orders multiply.
• Make cost per order the main tracking metric, not just overall ROI.

Phase 2 – Activation: raise in steps

• Raise in steps of 20–30% each time, holding for at least 3–5 days so the algorithm stabilises before the next increase.
• Track cost per order daily instead of looking only at overall ROI — a continuously rising cost per order is the signal to stop.
• Add new videos to GMV Max regularly so it doesn’t depend on a few old, fatigued videos.

Phase 3 – Scale: keep the rhythm at large scale

• Read results over a 7-day window to decide whether to keep raising or hold.
• Only raise further when orders grew faster than spend in the most recent week.
• Evaluate by total monthly profit, not by daily ROI.

Results

6 months (24/12/2024 – 22/06/2025)

Final 7 days of the period (15/06 – 22/06/2025)

At a scale of about 8 million VND in spend per day, orders grew faster than spend and cost per order fell — a signal that allows budget to keep rising step by step.

Analysis: where the growth came from

The GMV Max dashboard has no visitor count, so revenue is broken down as:

Revenue = Spend × (Orders per unit of spend) × Order value

Calculated from the dashboard data:

• Spend rose about 13.8 times.
• Order efficiency dropped to only about 35% of the previous period, as cost per order rose 186.75%.
• Order value fell about 6%, from ≈ 182,900 VND to ≈ 172,000 VND.

Multiplied together: 13.8 × 0.35 × 0.94 ≈ 4.5 times, matching the revenue increase of +353.75%.

So all of the growth came from a larger budget. Efficiency per dong fell, but not fast enough to cancel out the benefit of spending more.

The key question: was the extra spend worth it? Calculated from the dashboard (estimates):

• Extra spend ≈ 573 million VND.
• Extra revenue ≈ 13.67 billion VND.
• Marginal ROI ≈ 23.9, with cost per extra order ≈ 7,100 VND.

A marginal ROI of nearly 24 means each extra dong spent still brought in about 24 dong of revenue. For most consumer goods margins, this is spend that is very much worth it.

The weaker points:

• The final week’s ROI (19.23) is lower than the 6-month average (28.39). At higher spend, efficiency per dong keeps falling. This is a natural limit and needs monitoring to know when to stop.
• Cost per order nearly tripled. Without new videos, cost per order may keep rising as budget increases.
• Order value fell about 6%. This can be offset with bundles or gifts at an order-value threshold.

Lessons for store owners

• A high ROI on a small budget isn’t necessarily beneficial. Total profit is the ultimate yardstick.
• Scaling always has a “price”. Spend rising 13.8 times while revenue rises 4.5 times is normal.
• Calculate marginal ROI before deciding. Extra revenue divided by extra spend tells you whether the last dong is still profitable.
• Raise in steps of 20–30%, holding for 3–5 days. Don’t raise too sharply at once and destabilise the algorithm.
• Track cost per order daily. A continuously rising cost per order is the signal to stop.

Actual results depend on product, price, inventory, budget and platform policies. This is not a commitment to results for every client.