Marketplace ad management weekly cadence: the profit board behind active optimisation

Marketplace ad management usually has a weekly call. That does not mean it has a weekly operating cadence. A call can review what happened. A cadence decides what must change before the next euro is spent.

The named mistake is the checklist week. The agency checks budgets on Monday, trims a few bids on Tuesday, harvests search terms on Wednesday, sends a Friday update and calls it “active optimisation”. Useful? Sometimes. Sufficient for a brand spending €5K+ across Amazon, bol and MediaMarkt? Not even close. A checklist can keep the ad account tidy while the profit leak quietly moves somewhere else: a low-margin SKU, a stockout, a price match, a broken Buy Box, a delayed promotion, or a target that looks efficient only because last month’s organic demand is doing the heavy lifting.

The better model is a weekly profit cadence. It treats every week as a decision cycle with three locks: margin permission, stock permission and evidence permission. If a campaign cannot pass those locks, it does not get more budget just because ROAS looks polite.

This is especially important for the kind of marketplace accounts that are big enough to outsource but still close enough to the owner that every budget move matters. At €5K a month, you do not have unlimited testing room. You need a rhythm that catches waste quickly, protects winners properly and turns agency work into commercially visible decisions.

What competitors usually get right

The good marketplace advertising guides all agree on a few sensible habits. Daily checks matter. Weekly search term reviews matter. Bid changes should be based on data, not mood. Auto and broad campaigns are useful for discovery, exact campaigns are useful for control, and monthly or quarterly reviews should look beyond campaign-level ACOS.

Marketplace Valet describes an agency operating cadence as a rhythm that connects advertising, inventory, listing health and P&L accountability. That is a strong idea because it moves the conversation away from meetings and toward execution. Jarvio’s Amazon PPC SOP is also practical: daily budget checks, weekly search term work, monthly strategy review and quarterly deep dives. SellerSprite leans into weekly campaign workflow: search term mining, negative keywords, bid adjustments and scaling winners. BidX frames Amazon PPC in stages: setup, optimisation and scaling, with break-even calculations and automation. Podean’s Amazon Ads algorithm update adds a useful warning: Amazon has been pushing more discovery through auto and broad match, so operators need a sharper balance between exploration and efficiency.

What most of this content misses is the decision latency between the ad account and the business. A search term can be paused in five minutes. A margin rule may take five days to confirm. Stock data may sit in a separate system. MediaMarkt retail media performance may arrive in a different format from Amazon. bol Sponsored Products may look fine until LVB economics or return rates are added. The agency can “optimise weekly” and still be too late to protect profit.

That is the angle FiveX can own: the weekly cadence is not an optimisation checklist. It is a profit-permission board.

The weekly profit cadence in one page

A useful cadence has five stages. They can fit into a week, but the order matters more than the day names.

1. Monday: protect the account. Check overspend, underspend, stock, Buy Box, listing status, campaign budgets and promotions. No scaling decisions yet.
2. Tuesday: classify the spend. Split campaigns, targets and SKUs into protect, scale, learn, fix or stop. Do this with contribution margin and stock cover beside ad metrics.
3. Wednesday: make bid and search-term moves. Harvest, negate, trim, isolate and restructure only after the classification is clear.
4. Thursday: decide cross-marketplace moves. Move budget between Amazon, bol and MediaMarkt only where profit capacity exists, not where a platform dashboard looks loudest.
5. Friday: close the decision log. Record what changed, why it changed, what evidence triggered it and what will be reviewed next week.

That final step is where many managed accounts fail. If you cannot read last Friday’s decision log, this Tuesday’s bid change becomes theatre. The account may improve, but nobody can explain whether the improvement came from a better target, a promotion, a competitor stockout or simple luck.

Lock 1: margin permission

ROAS is not permission. ACOS is not permission. A campaign earns permission when the SKU can afford the click after marketplace commission, fulfilment, returns, discounts and the agency or software cost layer.

Take NorthSea Naturals, a fictional supplement brand selling a 2-pack magnesium spray on Amazon.nl. The SKU sells for €24.95. After VAT handling, Amazon referral fee, fulfilment, packaging, expected returns and landed product cost, the brand keeps €7.10 contribution margin before ads. That means a 28.5% break-even ACOS before any management fee. If the agency charges a €1,250 retainer on a €6,000 monthly account, the loaded break-even for the promoted SKU group is closer to 23% once you allocate service cost.

The Amazon campaign reports 4.0x ROAS, or 25% ACOS. In a normal weekly checklist, that campaign might be left alone. In a profit cadence, it becomes a yellow label: efficient enough for the platform, slightly too expensive for the business. The action is not “pause everything”. The action is sharper: keep exact brand defense at 18% ACOS, cap broad discovery at €25 per day, harvest converting “magnesium spray for sleep” queries into an exact lane, and stop spending on “muscle spray” until conversion rate improves.

This is where FiveX’s profitability dashboards help the operator avoid the classic ROAS trap. The ad platform knows click cost and attributed revenue. FiveX connects that to SKU margin, fees and fulfilment economics, so the weekly decision is based on what the SKU can actually afford.

Lock 2: stock permission

Advertising can create a beautiful problem: it sells the product you cannot replenish fast enough. That is not a win. It is ranking momentum rented with tomorrow’s stockout.

Consider BorealBaby, a fictional baby monitor accessory brand selling on bol.com in the Netherlands and Belgium. The hero SKU has 310 units available, sells 22 units per day organically and another 14 units per day when Sponsored Products are active. Replenishment arrives in 18 days. At the current combined velocity, stock cover is only 8.6 days. The bol campaign shows a neat 5.2x ROAS, but the SKU will run out before the purchase order lands.

A checklist week says: good ROAS, keep budget. A profit cadence says: protect the listing. Move the SKU from scale to protect. Drop non-brand discovery bids by 35%, keep brand and high-intent exact targets visible, shift €450 of weekly budget to a second SKU with 42 days of cover and a 31% contribution margin, and add a replenishment note to the Friday decision log.

FiveX inventory insights are useful here because marketplace advertising decisions need live stock context. The operator should not ask, “Which campaign has the best ROAS?” first. The better question is, “Which profitable SKU can still fulfil the demand we are about to buy?”

Lock 3: evidence permission

The third lock is the most uncomfortable one. A campaign is not automatically scalable because it converted last week. It needs enough evidence to justify the next budget move.

Evidence permission has three parts:

• Volume: enough clicks or orders to make the signal meaningful.
• Role: clarity on whether the campaign is protecting demand, discovering demand, launching a SKU, defending a category or clearing stock.
• Incremental logic: a reason to believe the spend created demand or protected demand that would otherwise be lost.

VoltEdge, a fictional electronics accessories seller, runs MediaMarkt retail media and Amazon Sponsored Products for a USB-C docking station. MediaMarkt spends €900 in a week and attributes €5,400 revenue. Amazon spends €1,800 and attributes €8,100. The platform view says Amazon should get more budget because it has more volume. The profit cadence asks for evidence by role.

MediaMarkt clicks are expensive, but the SKU has a 34% contribution margin, low returns and 55 days of cover. Amazon has higher volume, but only 19% margin after FBA fees and a rising return rate because shoppers are confusing laptop compatibility. The weekly decision is to add €300 to MediaMarkt category placements, hold Amazon budget flat, and send the Amazon listing team a compatibility-content task before scaling. That is not anti-Amazon. It is pro-profit.

This is where FiveX advertising automation and AI recommendations should be governed by permissions rather than left to chase surface metrics. Automation is excellent at making many small changes. The operator’s job is to define which changes are allowed.

The decision board: five labels every managed account needs

A weekly cadence becomes much easier when every SKU-campaign combination receives one of five labels.

The point is not to make the dashboard pretty. The point is to make trade-offs explicit. A bol campaign with 6x ROAS can still be fix if stock cover is nine days. An Amazon campaign with 2.8x ROAS can be protect if it defends a high-margin hero SKU from aggressive competitors. A MediaMarkt placement with weak last-click attribution can be learn if it supports a product launch with enough margin and stock.

What the weekly agency update should include

If you work with a marketplace advertising service, ask for a weekly update that separates reporting from decisions. The minimum useful version has six parts:

1. Spend movement: where budget moved by marketplace, campaign role and SKU group.
2. Profit permission changes: which SKUs changed label and why.
3. Search-term actions: harvested winners, negatives, watchlist terms and the margin logic behind them.
4. Operational blockers: stock, Buy Box, price, listing, review or return issues that limited ad decisions.
5. Automation rules touched: bid rules, budget caps, AI recommendations or exclusions changed during the week.
6. Next-week tests: the small number of decisions that need fresh evidence.

This structure avoids the “we optimised 47 bids” update. Nobody needs 47 bid changes in a management summary. They need to know which commercial bet changed.

How this changes the client-agency relationship

The weekly profit cadence is slightly less comfortable than a normal performance update. Good. Comfort is overrated when ad spend is leaking.

The brand has to provide cleaner margin, stock and promotion data. The agency has to explain not only what it changed, but what it refused to change. The operator has to admit when a campaign is waiting on content, price or fulfilment before advertising can do its job. That is healthier than pretending every growth problem can be solved with bids.

For Amazon, the cadence protects you from broad-match exploration that outruns proof. For bol, it keeps Sponsored Products tied to LVB economics, stock cover and product-level margin. For MediaMarkt, it forces retail media tests to earn budget even when reporting is less familiar than Amazon’s.

FiveX fits naturally into that operating model because the platform connects marketplace analytics, ad performance, profitability, inventory and AI recommendations in one place. The service conversation becomes much sharper: “This SKU is scale because margin is 32%, stock cover is 46 days and exact non-brand terms have 38 orders at 21% ACOS,” not “ROAS is up, so we added budget.” Tiny wording change. Massive business difference.

A simple 30-minute Friday review

If you want to start next week, do not rebuild the whole account. Start with a 30-minute Friday review:

• Pick the top 20 SKUs by ad spend across Amazon, bol and MediaMarkt.
• Add contribution margin, stock cover, return rate and campaign role beside each SKU.
• Assign protect, scale, learn, fix or stop.
• Write one sentence per label change: “Moved BorealBaby monitor clip from scale to protect because stock cover dropped below 10 days.”
• Choose three actions for Monday: one budget move, one search-term action and one operational blocker to resolve.

That small ritual will teach you more than another generic weekly dashboard. It also gives your agency a clearer mandate. They are not hired to touch buttons. They are hired to operate profitable marketplace demand under constraints.

The operator stance

Weekly optimisation is not wrong. It is just incomplete. Bids, budgets and negatives are the hands of the system. Margin, stock and evidence are the brain.

If your marketplace advertising service manages €5K+ per month, the weekly question should not be, “What did we optimise?” It should be, “Which spend earned permission, which spend lost permission, and what did we learn before the next euro left the account?”

That is the difference between an active agency and a useful one. One keeps the account moving. The other keeps the business moving in the right direction.