Most ad teams talk about stockouts too late.
The product is already unavailable, the campaign has already lost momentum, and someone is already asking whether Amazon will pause the ads automatically. Sometimes it will. Sometimes it will not. Sponsored Products often stop serving when the ASIN is not eligible. Sponsored Brands, DSP-style activity, retailer placements and broader retail media commitments can keep spending around the problem. bol and MediaMarkt add their own flavour: delivery promises, seller score pressure, retailer approval cycles and campaign budgets that do not politely wait for your next inbound shipment.
The named mistake is treating inventory as an emergency brake. The operator waits until stock is “low”, then slams spend down. That prevents some waste, but it also creates panic decisions: cut the best campaign, protect the wrong SKU, starve a profitable channel, or keep pushing because the report still shows nice ROAS.
My stance: every Advertentie Service managing roughly €5K or more per month across Amazon, bol and MediaMarkt needs an inventory runway. Not a stock dashboard. A spend-permission ladder that says what ads are allowed to do when a SKU has 28, 21, 14, 7 or 0 days of sellable stock left. The goal is not simply to avoid out-of-stock clicks. The goal is to slow demand before the business is forced into ugly trade-offs.
This matters because advertising is demand acceleration. If the SKU has healthy margin, stable offer eligibility and 60 days of cover, acceleration is useful. If it has 9 days of cover and the next container lands in three weeks, acceleration becomes a margin leak wearing a growth costume.
What the existing advice gets right
The public advice around Amazon ads and low inventory is useful. Optmyzr makes the key point clearly: Amazon does not save you from every out-of-stock scenario. Sponsored Products may stop when a product becomes ineligible, but Sponsored Brands, DSP and multi-ASIN campaigns can keep spending in ways the seller did not expect. Their rule-engine approach also highlights a practical reality: ad data and inventory data usually live in different places, so automation needs a shared data source before it can act.
Intentwise explains the same operational link from another angle. Inventory management is not separate from advertising management. If FBA inventory runs low, ad spend can accelerate the very stockout that later hurts organic rank, sales velocity and delivery experience. Their note about Vendor Central data lag is especially important for larger brands: if inventory is 48 hours late, your “real-time” ad decision is already yesterday’s decision.
Ad Badger adds a useful warning about pausing. Campaigns have history. Pausing everything too quickly can destroy learning and create more rebuilding work later. It also calls out Sponsored Brands with out-of-stock products as a special risk when those products were carrying most of the sales.
BidX’s Buy Box guide gives the other half of the problem. On Amazon, advertising permission is tied to offer strength. If price, fulfilment, seller performance or availability weakens the Buy Box, ads may stop spending or spend less predictably. In other words, the stock problem is also an offer problem.
What these pieces rarely do is turn inventory into a weekly budget decision across marketplaces. They explain when to pause. They do not give the operator a runway model for when to slow, shift, isolate, protect learning, or move spend from Amazon to bol or MediaMarkt before the crisis. That is the gap FiveX can own.
The operator view: stock is a budget permission signal
In a managed account, stock should not be a note in the margin of the weekly report. It should be one of the signals that decides whether the next euro may move.
A campaign can look healthy while it is quietly creating an operational problem. Imagine a kitchen brand selling a pan set on Amazon with 24% ACOS, 31% contribution margin and 11 days of FBA cover. The campaign is not “bad”. In isolation, it might deserve more budget. But if the next inbound shipment lands in 18 days, increasing spend is not optimisation. It is a decision to run out faster, lose sales velocity, risk ranking recovery costs and then spend again to rebuild demand.
The better question is not “should we pause?” It is: what level of demand are we still allowed to create?
That question turns inventory into a permission ladder. At 28+ days of stock, a profitable SKU may scale. At 21 days, it may keep harvest campaigns open but cap experiments. At 14 days, it may reduce conquest spend and protect only high-intent branded or exact traffic. At 7 days, it may move into stock-preservation mode. At 0 days, the issue becomes restart planning, not optimisation.
FiveX helps because the signals are not trapped in separate tabs. Inventory cover, SKU contribution margin, ad spend, channel revenue, repricing context and product-level performance can sit together. That lets the ad service see a campaign as a commercial decision, not just a line in Amazon Ads or bol Ads.
The inventory runway ladder
Here is the practical ladder I like for €5K+ marketplace ad accounts. Adjust the exact days by category velocity and replenishment reliability, but keep the principle: permissions should change before the stockout.
Green: 28+ days of sellable stock
The SKU has enough cover to absorb normal campaign learning. Scale is allowed if margin, offer strength and search evidence agree. This is where the team can run Sponsored Products expansion, bol category visibility, MediaMarkt retailer placements or competitor targeting. The rule is simple: growth experiments are allowed because fulfilment can support the demand.
Amber: 21 to 27 days
The SKU is still safe, but not carefree. Keep proven campaigns running. Reduce noisy exploration. Do not launch large new placements unless stock is already inbound and reliable. If Amazon is spending €90 per day and bol is spending €35 per day, the ad service should ask which channel has the cleaner contribution margin and faster replenishment path before adding another €25.
Orange: 14 to 20 days
This is where many teams are still celebrating performance. I would rather start slowing. Keep defensive and high-intent traffic. Cut broad match experiments, weak competitor targeting and placements that need time to learn. If a campaign cannot prove it creates profitable incremental demand within the remaining stock window, it should wait.
Red: 7 to 13 days
The account is now preserving stock. Ads may still run, but only with a clear reason: protect branded demand, sell through a seasonal batch before a planned replacement, or keep a strategic retailer relationship warm. Budget increases are blocked. AI bid recommendations should go into review instead of being applied blindly. If the SKU is a hero product, finance and operations should see the decision before advertising accelerates it further.
Black: 0 to 6 days
This is not normal optimisation. Pause, isolate or redirect spend depending on ad type, marketplace and substitute products. If a campaign promotes multiple ASINs, remove the weak SKU or change the landing route. If a retailer placement is already booked, negotiate a replacement SKU. If the product is already out of stock, document the restart rule now: when does spend reopen, at what daily cap, and which evidence window proves demand has recovered?
Example 1: DuneChef protects rank instead of chasing the last orders
DuneChef sells a stainless-steel pan set on Amazon.nl and bol. The Amazon campaign spends €82 per day at 23% ACOS. The SKU has 30% contribution margin after fees and fulfilment, so the campaign looks profitable. But FiveX inventory insights show 12 days of FBA cover and 19 days until the next inbound shipment is likely to become sellable.
The old move would be to keep spending because ACOS is below target. The inventory-runway move is different. The ad service cuts Amazon broad and competitor campaigns by 70%, keeps exact high-intent terms at €22 per day, and moves €18 per day to a bol SKU with 46 days of stock and similar margin. The team gives up roughly €420 of short-term Amazon attributed revenue in the week. It also avoids running out five days earlier and prevents a more expensive restart.
The important detail: the decision was not “Amazon bad, bol good”. It was “this Amazon SKU no longer has demand-acceleration permission, while the bol SKU does.”
Example 2: BrightBrew stops a MediaMarkt placement from becoming a stockout machine
BrightBrew sells espresso accessories through MediaMarkt marketplace and Amazon. A MediaMarkt retail media package is available for €1,400 across a weekend. The promoted milk frother has 18% net contribution margin, 16 days of stock cover and a supplier delay that could add another 10 days. In the MediaMarkt dashboard, the placement forecast looks attractive: 95,000 impressions and an expected 380 orders.
But 380 extra orders would consume 42% of available stock in three days. If the product sells through, Amazon loses the same SKU two weeks later, and the team will have to rebuild visibility on two channels. The ad service marks the placement hold, swaps in a bundle with 39 days of cover, and caps the original SKU’s Amazon campaign at €30 per day until inbound stock is confirmed.
That is the inventory runway doing its job. It does not reject retail media. It stops one attractive placement from creating a cross-channel fulfilment problem.
Example 3: NorthPeak uses low stock to clean up learning budget
NorthPeak runs Amazon Sponsored Products, bol Sponsored Products and a small MediaMarkt test for a portable power station. Monthly ad spend is €7,800. The hero SKU has 9 days of stock, 34% margin and a strong Buy Box position. The account also has €1,050 per month in experimental spend across broad keywords, competitor products and category placements.
Low stock forces a useful question: which learning still deserves to happen? The answer is not “pause everything”. NorthPeak keeps €12 per day on exact terms that defend profitable demand, pauses €28 per day of broad Amazon discovery, pauses €17 per day of bol category tests, and keeps a €10 per day MediaMarkt test because the retailer audience has higher average order value and the SKU has a substitute bundle available.
FiveX ad logs make that decision traceable. Three weeks later, when stock recovers, the team knows which tests were paused because of runway pressure, which were paused because they were weak, and which can restart without re-debating the whole account.
How to run the weekly inventory runway meeting
This meeting should take 20 minutes. If it takes an hour, the system is too complicated.
Start with every advertised SKU that has fewer than 28 days of sellable stock. Add contribution margin, current ad spend, channel split, open purchase orders, expected inbound date, offer status and campaign role. Then assign one of five labels: scale, hold, slow, preserve or stop.
Scale means stock is no longer the limiting factor. Hold means keep proven spend but do not add more. Slow means reduce learning and conquest spend. Preserve means only defensive or strategically necessary demand may remain. Stop means the SKU should not receive paid demand until stock, offer or fulfilment recovers.
The operator should also write down the reopen condition. “Restart when stock is back” is too vague. Better: “Restart broad Amazon campaign at €20 per day when sellable stock is above 21 days, Buy Box share is above 85%, and first 72 hours after restock show conversion within 15% of pre-stockout baseline.” That is specific enough to execute without a meeting.
This is where FiveX’s profitability dashboards and advertising automation become useful together. The dashboard shows which SKUs have margin and stock permission. Automation can then reduce bids, cap budgets or hold recommendations when the runway label changes. The human still owns the commercial rule; the system makes sure the rule does not get forgotten on a busy Tuesday.
The trade-off: you will sometimes slow a profitable campaign
This is the uncomfortable part. An inventory runway will occasionally reduce spend on a campaign that looks profitable today.
That feels wrong if the ad team is judged only on ACOS or ROAS. It is correct if the business is judged on contribution margin, rank recovery cost, customer delivery experience and channel availability. The last ten units of a popular product often do not need paid help. They need protection from overeager advertising.
The operator voice here is simple: do not let the auction decide how fast your warehouse empties.
Marketplace ads should create demand the business can fulfil profitably. If stock cover, offer eligibility or replenishment reliability says the SKU cannot handle more demand, the ad service should slow down before the marketplace forces the issue. That is not conservative. It is commercially mature.
Final thought
Pausing ads after a stockout is housekeeping. Building an inventory runway is management.
For Amazon, bol and MediaMarkt accounts above €5K monthly spend, the difference is material. The first reacts to waste after it appears. The second prevents demand from outrunning margin, stock and operational trust.
If your marketplace ad service cannot show which SKUs are allowed to accelerate, which must slow down and which are waiting for inventory permission, it is not managing demand yet. It is managing campaigns. Those are not the same job.