On 27 October 2026, Meta’s personal AI agent Muse went live in Singapore. The Meta newsroom post, dated 28 October 2026 UTC, carries the line that matters to you: you can ask Muse to research products and compare prices before buying on Shopee or Lazada. Shopee is one of eight connected local apps. The consumer’s comparison work is now done by a machine.
What this means for your margin is narrower than the headlines suggest. It does not mean prices collapse. It means the 5% gap you used to hide behind a better product photo is now visible before checkout. If your price sits 5% above a competitor and your margin on that SKU is 12%, the competitor’s price is not a nuisance — it is more than a third of your profit. The agent does not need to be right about your product. It only needs to be right about the number.
Turn it into arithmetic
Assume, for illustration only, a Singapore order at S$20.00, a landed cost of S$14.00, and therefore a gross margin of S$6.00, or 30%. This is our assumption to make the arithmetic visible. It is not a rate from your store and not a Shopee statistic. Substitute your own cost.
The formula is one line:
Margin points lost = (Competitor price gap % × Your selling price) ÷ Your gross profit
Run it on the S$20 order. A competitor lists the same item at S$19.00, a 5% gap. You match at S$19.00. You give up S$1.00. On S$6.00 of gross profit, that is 16.7 points of margin — your 30% becomes 13.3%. A 5% price cut cost you more than half your margin. That is the whole story, and it is why the gap percentage is the wrong number to watch. The right number is the gap divided by your gross profit.
| Line | Before the 5% gap | After matching a 5% gap |
|---|---|---|
| Selling price | S$20.00 | S$19.00 |
| Landed cost | S$14.00 | S$14.00 |
| Gross profit | S$6.00 | S$5.00 |
| Gross margin | 30.0% | 26.3% |
| Margin points lost | — | 3.7 points |
| Profit lost vs. before | — | S$1.00 per order |
Wait — the two calculations above disagree, and that is the point of the table. The first line treats the S$1.00 as a share of the original S$6.00 profit: 16.7%. The table treats it as a share of the new S$19.00 price: 3.7 margin points. Both are correct; they answer different questions. Use the profit-share number when you decide whether to follow. Use the margin-point number when you report to your boss. Sellers who quote only the second number talk themselves into following every gap, because 3.7 points sounds cheap.
Now push the same order to a thin-margin SKU. Same S$20.00 price, landed cost S$17.00, gross profit S$3.00, margin 15%. A competitor at S$19.00 is a 5% gap. Matching costs S$1.00, which is 33.3% of your gross profit. Your 15% margin becomes 10.5%. On a 15% SKU, a 5% gap is a third of your profit. On a 30% SKU, the same 5% gap is a sixth. The gap is identical; the damage is not.
That gives you a decision rule you can apply without a spreadsheet. Divide the competitor’s gap percentage by your gross margin percentage. If the result is above 0.33, do not follow on price — hold and defend on rating or bundle. If it is below 0.20, following is cheap enough to consider. Between 0.20 and 0.33, it depends on sales increment, which is the third input in smart follow: follow, hold or raise. On the 30% SKU, 5% ÷ 30% = 0.17, so following is cheap. On the 15% SKU, 5% ÷ 15% = 0.33, so you are at the edge. That is the arithmetic the agent never shows the consumer, and it is the only arithmetic that matters to you.
The counter-intuitive turn
The common practice that stops working here is the daily manual price check — open the seller centre, scan three competitors, adjust. It fails not because it is lazy but because it is sampled. Two checks a day on a fast-moving SKU is two data points against a consumer agent that reads the listing at the moment of purchase. The failure is arithmetic, not attitudinal.
Suppose a competitor moves price four times in a day, in S$0.50 steps from S$20.00 down to S$18.50. You check at 09:00 and 15:00. You see at most two of those four moves, and if the moves happen between your checks you see none of them. The average gap you observe is not the average gap that exists. You are pricing against a sample, and the consumer is pricing against the population. On a S$20 order with S$6.00 gross profit, missing one S$0.50 step is 8.3% of your profit you never saw move.
The second practice that fails is following every gap you do see. If you match each observed 5% gap on the 15% SKU, you lose 33.3% of gross profit per order to defend a price position that the agent may not even have surfaced. The seller who follows every gap is not defending margin; they are donating it. The seller who ignores every gap is not protecting margin either — they are waiting to be filtered. The decision rule above sits between those two failures, and it needs a third input: sales increment. A gap you did not follow, on a SKU whose unit sales rose 20% week on week, was the right call. A gap you did not follow, on a SKU whose sales fell 15%, was a mistake you can now measure.
This is why the cadence matters more than the cleverness. How the check schedule works explains the fixed schedule: two checks a day on Free, four on Pro. Four checks a day is still a sample. The honest response is not to pretend it is continuous — it is to make each check produce a decision instead of a number.
Where we sit in this
PriceMinder monitors Shopee only, across seven markets: ID, VN, TH, PH, MY, SG and BR. Free gives you 5 products, 2 checks a day, a 7-day price trend and alerts on desktop and Feishu. Pro gives you 50 products, 4 checks a day, a 30-day trend, Smart Follow’s three-input decision with a 14-day review, Telegram and Slack alerts, and unlimited MCP tool calls against the 10-tool set. The honest limitation: we do not offer instant checks, there is no auto price change, and other platforms are out of scope — Muse connects Shopee and Lazada, but we monitor Shopee only. That is the honest choice because a fixed four-check schedule with a written decision rule produces an auditable trail you can review in 14 days, while a continuous feed produces alerts you cannot act on and a price you changed without remembering why. If you want the plan detail, it is on pricing and plans.
Numbered action checklist
-
Today — pull your top 10 SKUs and write down two numbers each: selling price and gross profit in dollars. Not margin percentage, gross profit. On a S$20 order with S$6.00 profit, write 6.00. You cannot apply the rule without the dollar figure. Ten SKUs, twenty numbers, one sitting.
-
Today — compute the gap-to-margin ratio for each of the 10. Competitor gap % ÷ your gross margin %. Flag anything above 0.33 as do-not-follow. On the 15% SKU above, 5% ÷ 15% = 0.33, so it is flagged. Expect three or four of your ten to land there.
-
This week — set the two thresholds per flagged SKU: a floor that protects gross profit and a ceiling that stops you overpricing into the agent’s filter. The floor is your cost plus the profit you will not go below. The ceiling is the highest price at which your gap to the cheapest credible competitor stays under 5%. Write both down; a threshold you did not write down is a mood.
-
This week — load the competitor list into the monitor and set the cadence. Free covers 5 products at 2 checks a day; Pro covers 50 at 4 checks a day. If you sell in Singapore and the agent is live in your category, pick Pro for the SKUs you flagged in step 2. The 7-day trend on Free is too short to see a competitor’s promo cycle; 30 days on Pro is the minimum for a monthly pattern.
-
This month — run one 14-day Smart Follow review on the three SKUs where you chose Hold over Follow. The review exists to answer one question: did unit sales rise or fall after you held? A hold that cost 15% of unit sales was wrong; a hold that cost nothing was right. Fourteen days is the review window because it covers two weekly cycles.
-
This month — add a rating check to every SKU where you are holding a price above a competitor. If your rating is not at least 0.3 stars above theirs, holding a 5% gap is a bet you will lose on the agent’s comparison screen. Rating is the second input in the decision, not a consolation prize.
-
This month — write the rule into your SOP in one sentence: “If gap ÷ margin > 0.33, hold and defend on rating or bundle; if < 0.20, follow; between, review sales increment at the next check.” One sentence, applied at each scheduled check, beats a dashboard nobody reads. The repricing math post shows the same division applied to a tariff, if you want a second worked case.
FAQ
Does Muse actually compare prices on Shopee? Yes. Meta’s own newsroom release on 28 October 2026 (UTC) states you can ask Muse to research products and compare prices before buying on Shopee or Lazada. Singapore is the launch market, and Shopee is one of eight connected local apps. Local coverage reported 27 October because of the time zone.
How often does PriceMinder check competitor prices? On a fixed schedule, not continuously. Free accounts get two checks per day, morning and afternoon. Pro accounts get four: morning, afternoon, evening and night. There is no instant check and no auto price change. The schedule is documented at how the check schedule works.
What is the 5% gap rule? If your price is more than 5% above a competitor on an order where your margin is under 15%, the gap is larger than a third of your profit. Follow or hold only when the gap divided by your gross margin stays under 0.33. The worked example shows this on a S$20 order with S$6.00 profit.
Can I monitor Lazada or other platforms with PriceMinder? No. PriceMinder covers Shopee only, in seven markets: ID, VN, TH, PH, MY, SG and BR. Other platforms are out of scope. Muse connects Shopee and Lazada, but our tool does not monitor Lazada, and we do not claim to. The seller centre remains the authoritative source for your own listing and fee data; our numbers are a monitoring input, not a platform statement.