Marketplace ad retainers often renew for the wrong reason: nothing exploded.
The campaigns kept spending. The weekly reports arrived. ACOS moved a little. Search terms were harvested. A few bids went up, a few went down, and everyone survived the quarter. That can feel like progress, especially when an Amazon, bol or MediaMarkt account has enough moving parts to make silence look like stability.
But stability is not the same as earned renewal.
The named mistake I see is renewing the marketplace ad service because activity was visible instead of because decision quality improved. The agency shows 143 bid changes, 68 negative keywords, four campaign restructures and a dashboard with nicer trend lines. The brand signs another quarter. Nobody asks the harder operator question: did the retainer make better calls about which SKUs deserved spend, which campaigns should have waited, and which marketplace should receive the next euro?
My stance: €5K+ managed marketplace ad accounts need a renewal evidence room. Not a polite end-of-quarter presentation. A decision room that compares the agency’s work against profit permission, stock discipline, offer quality, learning evidence and next-quarter budget rights. If the retainer cannot prove better decisions, it should not automatically renew. It should be resized, rescoped, paused or replaced.
This is especially important for brands selling across Amazon, bol.com and MediaMarkt in NL/BE. Platform reports reward local optimisation. Renewal decisions need portfolio judgement. A campaign can improve ACOS while the SKU runs out of stock. A bol.com push can win cheap sales while MediaMarkt carries the healthier margin. An Amazon defence campaign can look necessary until you see that branded demand stayed stable during a controlled budget cap. Renewal should judge that whole chain.
What the public advice gets right
The research landscape is useful, but incomplete. Podean’s Amazon agency guidance makes a strong point: Amazon agencies are not just media buyers. The best ones understand retail operations, inventory, product pages, chargebacks, replenishment and the many Amazon teams a brand has to navigate. That is the correct starting point. Amazon advertising is never only advertising.
BidX’s pricing and positioning pages show another important reality: the market has split into self-service software, managed platform and managed service models. Their public pricing makes the trade-off visible. A seller can pay for automation and a monthly strategy layer, or pay materially more for a team that handles research, strategy and operations. The shape of the service matters.
Daniks.AI frames the agency-versus-software decision with useful arithmetic. A $2,000 retainer on $6,000 ad spend can be mathematically unfair even when the agency improves ACOS. RedHen Labs goes one step sharper: agency fees should be compared with the contribution advertising actually creates, not with revenue. That is the right instinct for renewal conversations.
Reddit threads, seller forums and YouTube audit content add the emotional layer. Sellers rarely complain that an agency never touched the account. They complain that they cannot tell what changed, why it changed, whether it improved real profit, or whether the agency is hiding behind platform metrics. The fear is not inactivity. The fear is paying for motion that nobody can connect to commercial judgement.
The gap: renewal is treated like procurement, not a trading decision
Most renewal conversations still sound like vendor management. Was the agency responsive? Did reports arrive? Did the relationship feel good? Is the fee acceptable? Should we negotiate the retainer?
Those questions matter, but they are not enough for marketplace advertising. A €7,500 monthly ad account is a trading system. Every week it decides which products get demand, which products are protected, which terms are allowed to learn, which marketplaces receive budget and which risks are accepted. Renewing the service means renewing those decision rights.
That is why the renewal evidence room should not start with a slide deck. It should start with a ledger.
The five proofs every renewal evidence room needs
A good renewal pack has five kinds of proof. If one is missing, the conversation becomes too easy.
1. Profit proof: did spend move toward contribution, not just ROAS?
Platform ROAS is not enough. The renewal room should show spend, ad-attributed revenue, total revenue, contribution margin, return reserve and TACoS by SKU and marketplace. The key question is simple: did the operator move money toward the products where paid demand could still create profit?
This is where FiveX’s profitability dashboards earn their keep. When Amazon, bol.com and MediaMarkt orders, fees, purchase costs, fulfilment costs and ad spend live in one view, the renewal conversation stops being “ACOS was 24%” and becomes “this campaign created €1,840 contribution after returns and stock constraints”. Much better coffee-table conversation. Much harder to fake.
2. Constraint proof: did the agency respect stock, offer and margin vetoes?
Great ad management is often the art of not spending. The renewal evidence room should list every time the agency capped, paused or refused scale because a SKU lacked permission: low stock cover, weak Buy Box, worse delivery promise, stale price, poor review readiness, expired margin file or return spike.
A retainer deserves renewal when it protected the business from attractive but unsafe clicks. If the agency only reports what it scaled, it is hiding half the job.
3. Learning proof: what did the budget teach that will change next quarter?
Learning budget should not evaporate into “we tested”. The room needs a register of experiments, hypotheses, spend, result, decision and expiry date. Did Amazon Sponsored Products prove a new non-brand cluster? Did bol Sponsored Products show that Belgium converts differently from the Netherlands? Did MediaMarkt retail media prove that a bundle needs a different price before traffic scales?
FiveX advertising reports can pull the weekly evidence together, but the operator still has to write the judgement. Automation can surface the pattern. A human needs to decide whether the pattern deserves next-quarter money.
4. Change proof: can every important move be explained?
Renewal should include the top 20 decisions, not the top 200 clicks in the change log. For each important move, capture the trigger, data used, action, expected outcome, review date and result. Bid increases without review dates are just tiny promises scattered across the account.
FiveX change logs and Ads AI recommendation history help here. They make it possible to separate human judgement from automation, and to see whether recommendations were accepted, edited or rejected. That distinction matters when you renew a service: you are not only renewing the software layer, you are renewing the quality of the operator sitting above it.
5. Next-euro proof: where should the next quarter’s budget go?
The renewal meeting should end with budget rights, not a vague optimism slide. Which SKUs can scale? Which campaigns stay capped? Which marketplace gets incremental budget first? Which experiments are retired? Which ones graduate from learning to working spend?
If the agency cannot answer “where should the next €1,000 go and why?”, the retainer has not produced enough decision evidence.
Named example 1: NovaNest Home and the ACOS renewal trap
NovaNest Home spends €8,400 per month across Amazon.nl and bol.com on storage baskets and small home organisers. The agency renewal deck looks decent: ACOS improved from 31% to 26%, ad-attributed revenue rose from €27,100 to €32,300, and 74 negative keywords were added.
The renewal evidence room changes the conclusion. FiveX shows that the hero basket has a loaded break-even ACOS of 22% after purchase cost, bol commission, FBA/LVB fees and a 9% return reserve. Amazon ads improved, yes, but the agency scaled the SKU while stock cover fell from 34 days to 11 days. At the same time, a bol.com long-tail campaign on a second organiser held 19% ACOS against a 31% break-even ceiling and still had 58 days of stock.
The decision is not “fire the agency”. It is “renew with changed rights”. Amazon hero campaigns move from Scale to Hold. The bol long-tail campaign gets a €1,200 quarterly budget release. The agency’s next-quarter SLA now requires stock-cover vetoes before bid increases above 15%. The retainer stays, but the permission model changes.
Named example 2: BrightBrew and the busy-work renewal
BrightBrew sells espresso accessories on Amazon and MediaMarkt. Monthly ad spend is €6,200. The agency reports 212 bid changes, 37 campaign edits and a clean weekly meeting cadence. Everyone feels looked after.
The renewal room asks a less comfortable question: which changes mattered? Only six decisions moved more than €150 in spend exposure. Three were small bid cuts on targets already below €20 monthly spend. Two were campaign renames. One mattered: MediaMarkt budget was capped after conversion dropped when the delivery promise slipped from two days to five.
That single good decision saved roughly €620 in wasted spend. Useful. But the rest of the retainer was maintenance theatre. BrightBrew does not cancel immediately; it resizes. The next quarter moves from full managed service to a lighter managed platform model, with FiveX automation handling bid hygiene and the operator reserved for weekly budget court, offer changes and new launches. The brand keeps expertise without paying premium rates for microscopic motion.
Named example 3: LumiClean and the renewal that should expand
LumiClean sells premium cleaning devices across bol.com NL/BE and Amazon.de. Monthly ad spend is €11,500. The agency asks to expand scope to MediaMarkt retail media.
The evidence room supports the expansion. Over the quarter, the agency cut €1,900 from a branded Amazon defence campaign after a two-week holdout showed stable total sales. It shifted €2,400 into bol.com Belgium search terms with 28% higher contribution per order than the Dutch equivalent. It paused a MediaMarkt-ready test until the product page gained warranty clarity and enough stock to cover four weeks of demand.
That is not just optimisation. That is portfolio judgement. LumiClean renews, adds MediaMarkt scope, and sets a rule: the first €3,000 of MediaMarkt spend is released in three tranches only after margin, stock and offer evidence pass. This is the kind of retainer you want to grow.
A practical renewal scorecard
Use a simple 25-point scorecard before the contract rolls over:
- Profit evidence, 0–5: Can the agency show contribution margin by SKU and marketplace, not just ACOS?
- Constraint discipline, 0–5: Did it respect stock, margin, offer and review vetoes?
- Decision clarity, 0–5: Are the important changes documented with triggers, owners and review dates?
- Learning quality, 0–5: Did tests produce reusable budget decisions?
- Next-quarter plan, 0–5: Is there a clear next-euro allocation across Amazon, bol and MediaMarkt?
Score 20–25: renew and consider expanding scope. Score 15–19: renew with tighter permissions and a specific 30-day improvement plan. Score 10–14: resize or run a proof month with restricted budget rights. Below 10: do not renew the same model. You may still need help, but you need a different operating layer.
How FiveX supports the room without replacing judgement
FiveX does not make the renewal decision for you. That would be too easy, and a little suspicious. What it does is make the evidence hard to avoid.
Profitability dashboards connect ad spend to SKU-level margin, fees, fulfilment and returns. Stock insights show whether a campaign had permission to scale. Product groups let the operator compare hero SKUs, clearance items, launch products and margin-sensitive ranges without rebuilding spreadsheets. Advertising reports turn Amazon, bol and MediaMarkt performance into a common language. Ads AI and automation rules help with bid hygiene, while logs preserve what changed and why.
That combination is exactly what renewal needs: not more screenshots, but a cleaner argument about whether the retainer improved decisions.
The operator rule
Here is the rule I would use before any marketplace ad retainer renews:
If the agency can only show what it did, the renewal is not ready. If it can show which profitable decisions the business would not have made without it, the renewal deserves a serious yes.
That difference sounds small. It is not. One renews activity. The other renews judgement. In marketplace advertising, judgement is the only part worth paying premium rates for.