A generalist marketing agency can be a perfectly sensible place to start. They know your brand, your website, your Google Ads account, your creative calendar and the weekly campaign rhythm. If marketplace ads are still an experiment, that shared context is useful.
But there is a moment where Amazon, bol and MediaMarkt stop behaving like another paid media channel and start behaving like an operating system. Bids depend on Buy Box health. Sponsored Products performance depends on stock cover. ROAS can look clean while fulfilment fees, returns and price pressure quietly eat the profit. At that point, the question is no longer “can our agency manage campaigns?” It becomes: can our agency decide which SKUs deserve spend today?
The named mistake I see is the website-agency hand-me-down. A brand gives marketplace advertising to the team that already manages Google, Meta, Shopify CRO or hosting because it feels efficient. Nobody is being lazy. It is just tidy. One agency, one call, one invoice. Lovely. Then the account reaches €5,000 to €20,000 monthly marketplace spend and the cracks appear: the ad team optimizes CPC, while operations fights stockouts; finance knows SKU margin, but the campaign manager only sees ACOS; the marketplace manager knows bol LVB economics changed, but the media report still celebrates revenue.
My stance: for brands selling across Amazon, bol.com and MediaMarkt in NL/BE, marketplace advertising should move from generalist support to specialist operation when three things are true: spend is material, SKU profitability varies, and marketplace operations affect ad performance. That often starts around €5K monthly spend, not because €4,999 is magic, but because the cost of wrong decisions becomes visible.
This guide explains where generalist agencies still help, where specialist marketplace ad management pays for itself, and how to make the handover without turning the account into a reporting soap opera.
What competitor content gets right
The research landscape is useful, but incomplete. Podean positions marketplace management as a full-funnel discipline: strategy, media, SEO, A+ content, DSP, analytics, retail operations and global expansion. Their strongest point is that marketplace growth is not just campaign management. They also mention non-media signals like out-of-stock data, price changes and merchandising tactics informing retail media decisions. That is exactly the direction mature advertisers need.
BidX explains the software side well. Its agency material focuses on campaign creation, bid automation, dashboards, white-label reporting, stock-aware scheduling, BSR and keyword ranks. That is valuable for agencies handling many Amazon accounts because speed and audit trails matter. The missing piece is the commercial decision layer: when should the tool refuse spend because margin, inventory or marketplace role says “not today”?
JumpFly describes marketplace ads as one piece of a broader ecommerce puzzle and points out that brands may need help expanding beyond Amazon into other marketplaces. Good point. But most service pages stay broad: audits, campaign management, expansion recommendations and expert support. They rarely show how a €6,000 bol budget, a €12,000 Amazon budget and a €4,000 MediaMarkt test should compete for the same next euro.
SellerApp and many Amazon PPC agency pages do a good job listing the tactical jobs: keyword research, Sponsored Products, Sponsored Brands, Sponsored Display, negative targeting, bid management and ACoS reduction. Pricing guides from agencies such as Olifant Digital also make a useful point: the fee model matters. A retainer, percentage of spend or hybrid model changes incentives.
Reddit threads add the uncomfortable operator truth. Sellers worry that percentage-of-spend fees can reward higher spend rather than better profit. Others point out that low ad spend may not fund enough senior attention. Both are fair. The service model must be profitable for the agency and commercially clean for the seller. Pretending otherwise is how bad retainers are born.
What most content misses is the transition moment. It talks about choosing an Amazon PPC agency or buying automation software, but not about the messy middle: the brand already has a generalist agency, marketplace spend is now meaningful, and nobody wants to break what works. That is the angle FiveX can own: the handover from campaign management to marketplace operating control.
The real difference: channel manager vs SKU operator
A generalist agency usually thinks in channels. Google Search has budgets, campaigns, keywords and landing pages. Meta has audiences, creatives and funnels. Marketplace advertising has those elements too, which is why the channel looks deceptively familiar.
The specialist sees a different object: the SKU. Not the campaign. Not the ad group. The SKU.
For a marketplace ad operator, every product has a permission status. Can it scale? Should it defend rank only? Is it out of stock risk? Does it lose money after returns? Is bol more profitable than Amazon this week? Is MediaMarkt useful for visibility but not yet ready for scale? The answer changes daily because marketplace conditions change daily.
That is the core trade-off. Generalists bring brand and media breadth. Specialists bring marketplace-specific refusal power. The valuable specialist does not just find growth. They also say no to tempting spend.
The €5K threshold: why small mistakes start becoming expensive
At €800 monthly spend, a messy campaign structure is annoying. At €8,000, it is a profit leak with a dashboard. The account has enough volume to make bad assumptions costly, but often not enough headcount to build a full internal retail media team.
This is where many NL/BE ecommerce brands sit. They have 30 to 300 SKUs, sell through Amazon and bol, maybe test MediaMarkt, and spend somewhere between €5K and €50K per month across marketplaces. The founders or ecommerce lead can still name the important products, but cannot manually connect every bid decision to margin, stock, returns and price position.
The specialist agency earns its place when it changes the weekly conversation from:
- “ACOS improved from 24% to 20%.”
- “Spend was 92% of budget.”
- “Top campaign grew revenue by 18%.”
to:
- “We moved €1,400 from a low-margin Amazon defence campaign into three bol SKUs with 31% contribution margin and 22 days of stock.”
- “We capped MediaMarkt spend because the advertised dishwasher accessory had 9% net margin after returns.”
- “We protected the hero ASIN, but did not scale it because inbound FBA stock arrives in six days.”
That is a very different operating rhythm. It is also where FiveX fits naturally: the platform connects marketplace, advertising, inventory and financial data so the service team can see whether spend is commercially allowed before pushing more budget into the auction.
Named example 1: the Amazon hero SKU that should not scale yet
Imagine a Dutch electronics brand selling a USB-C docking station on Amazon.nl. The SKU sells for €79.95. After referral fees, fulfilment, payment costs and COGS, retained contribution margin before ads is €18.40 per unit, or 23%.
The generalist view looks promising. Sponsored Products spent €3,200 last month, generated €15,800 in attributed sales and reported 20.3% ACOS. The target ACOS is 24%, so the obvious recommendation is to raise budget.
The marketplace operator pauses. FiveX shows only eight days of FBA stock cover, Buy Box ownership dropped from 96% to 82% because a reseller undercut the price twice, and the parent SKU has a 13% return rate on one variation. If the team scales now, ads accelerate a stockout and hand organic rank to competitors.
The better move is boring but profitable: cap spend at €900 for the week, protect exact-match terms, pause broad discovery, trigger a price and reseller review, and reopen scale only when stock cover passes 21 days. The campaign did not fail. The SKU was temporarily not allowed to grow.
Named example 2: the bol LVB product where ROAS lies nicely
Now take a Belgian home brand selling lunch boxes on bol.com. The SKU sells for €24.99. Bol Sponsored Products delivers a 7.1 ROAS on €1,100 monthly spend. On a normal paid media report, that looks healthy.
But the operator view includes LVB costs, returns, marketplace commission, packaging and the real pick-pack economics. After all costs, the product keeps €4.20 contribution before ads. A click costs €0.38, conversion rate is 9.5%, so the ad cost per order is roughly €4.00. That leaves €0.20 before overhead. One small return-rate movement wipes it out.
A generalist may keep scaling because ROAS is above target. A specialist splits the campaign into roles. Exact terms for the brand and best converting product terms stay live because they defend rank. Generic “lunch box kids” traffic gets a hard CPC ceiling of €0.24 unless conversion improves. Budget moves to a stainless-steel bundle with €8.70 contribution margin and enough stock for four weeks.
This is not anti-growth. It is pro-grown-up growth. In FiveX, that decision is easier because SKU-level contribution margin, bol performance and ad spend sit in the same view instead of three exports and one heroic spreadsheet.
Named example 3: the MediaMarkt test that needs a learning budget, not a vanity budget
MediaMarkt is often different from Amazon or bol because the product set, shopper intent and category competition are narrower. Suppose a small appliance brand wants to test retail media for an air fryer accessory range. The team proposes €5,000 for the month because “that is our test budget.”
The specialist asks a less comfortable question: which decision will €5,000 buy? If the range has only 11 active SKUs, two hero products, limited reviews and a 14-day stock cover on the best seller, a full-scale test may mostly buy noise.
A better plan could be €1,800 over three weeks: €1,200 on the two SKUs with at least 28% contribution margin, €400 on category discovery, and €200 reserved for search term validation. The success metric is not platform ROAS alone. It is whether the two hero SKUs can acquire orders below €6 ad cost per order while maintaining price position and stock cover above 18 days.
If they pass, budget moves to €4,000 the next month. If they fail, the team fixes content, pricing or assortment before buying more traffic. FiveX AI recommendations can flag this pattern: do not increase MediaMarkt spend until conversion and margin thresholds are met. Very unglamorous. Very useful.
Five signs you have outgrown generalist marketplace ad management
1. Your agency reports ACOS, but finance asks for contribution margin
ACOS is useful, but it is not a P&L. If the weekly report cannot show break-even ACOS by SKU, campaign role and marketplace, the team is steering with one eye closed.
2. Stockouts surprise the ad team
If campaigns are still scaling products with less than two weeks of stock, advertising and operations are not connected. Marketplace ads should slow down before inventory creates the crisis, not after.
3. bol, Amazon and MediaMarkt each get separate reports
Separate platform reports make every channel look important. A specialist compares them by profit capacity. Sometimes the next €1,000 belongs to bol. Sometimes Amazon should defend only. Sometimes MediaMarkt needs a small learning budget rather than scale.
4. Your fee model rewards spend without a profit check
A percentage-of-spend fee is not automatically bad, but it needs guardrails. The agency should not earn more simply because wasted spend increased. Tie reviews to retained contribution, TACoS movement, stock-aware decisions and documented budget shifts.
5. Nobody owns the “do not advertise” list
Every marketplace account needs a list of SKUs that are temporarily blocked from scaling: low margin, low stock, high returns, weak price position, poor content or unstable Buy Box. If nobody owns that list, spend will find the weakest products eventually. It always does. Tiny menace, big invoice.
How to hand over without breaking the account
The safest transition is not a dramatic agency divorce. It is a controlled migration of decisions.
Week 1: map the current reality. Export campaigns, search terms, SKU sales, COGS, marketplace fees, stock cover, returns and price position. Label every campaign by role: defend, grow, harvest, test or waste review.
Week 2: build SKU permission rules. Define minimum contribution margin, stock cover, Buy Box or offer health, return-rate limits and break-even ACOS by product group. In FiveX, these rules can be monitored through profitability dashboards and alerts rather than rebuilt manually every Monday.
Week 3: change budget allocation, not everything at once. Keep winning exact campaigns stable. Reduce spend where SKU economics fail. Move a controlled amount into higher-permission SKUs. Document the reason for every meaningful shift.
Week 4: review profit, not just media efficiency. Compare ad spend, attributed sales, organic movement, TACoS, retained contribution, stock movement and returns. The goal is not to make the prettiest campaign report. The goal is to prove that the new operating model makes better commercial decisions.
What a specialist marketplace advertising service should deliver
If you are evaluating an Advertentie Service partner for Amazon, bol or MediaMarkt, ask for more than campaign maintenance. The service should include:
- SKU-level margin and break-even ACOS logic before scaling campaigns.
- Weekly budget moves across marketplaces, not just within one ad console.
- Stock-cover and Buy Box checks before bid increases.
- Clear campaign roles: defend, grow, harvest, test and stop.
- Search term governance, negative targeting and keyword harvesting.
- Retail media reporting that explains decisions, not only outcomes.
- A shared “do not scale” list owned by advertising, operations and finance.
FiveX’s managed marketplace advertising service is built around that operating model. The human team handles the marketplace ad work; the platform supplies the cross-marketplace profit, inventory and advertising context; AI recommendations help surface where budget should move or stop. That combination matters because marketplace advertising is too operational to run on media metrics alone.
The bottom line
A generalist agency is not the villain. Many do good work, especially while marketplace ads are small, brand context matters more than SKU complexity, and the account needs basic structure.
But once Amazon, bol and MediaMarkt spend becomes material, the job changes. You are no longer buying campaign management. You are buying commercial judgement under marketplace constraints.
The best operator knows when to scale, when to defend, when to test and when to refuse spend. That last one is the difference between an ad manager and a marketplace advertising specialist.
If your monthly marketplace spend is around €5K or more and the weekly report still cannot explain spend by SKU margin, stock cover and channel role, it is probably time to upgrade the operating model. Not because specialists sound fancier. Because the auction is already charging you for every blind spot.