Marketplace agencies are very good at managing the work that is visible. The client with a Slack thread gets a reply. The account with a reporting call gets a deck. The campaign with a red ACOS cell gets a bid change. The urgent listing issue gets pulled into the stand-up. None of that is wrong. It is also not enough.
The dangerous work is often the work that does not shout. A Walmart client has 13 days of cover on the only profitable variant. An Amazon client is spending $4,800 this month on campaigns that still look efficient because refunds have not landed. A Kaufland assortment has 312 suppressed offers after a taxonomy change, but the account manager is busy rebuilding a QBR for a louder client. A TikTok Shop test created demand for a SKU that Amazon is about to stock out on. The agency is busy. The portfolio is not being steered.
The named mistake I see is letting client volume decide agency attention. The biggest client, loudest stakeholder or nearest meeting gets the senior brainpower, while the account with the highest commercial exposure quietly waits. For a marketplace agency with five or more employees, that is how good teams become reactive. They do not lose control because they lack dashboards. They lose control because they do not have a shared rule for which client deserves attention first.
My stance: every marketplace agency needs a portfolio risk board. Not a nicer client dashboard. Not another task list. A daily operating layer that ranks clients, channels and SKUs by profit exposure, decision urgency and agency capacity before work is assigned. The board answers one practical question: “If we only have six focused hours today, where do those hours protect the most client profit and agency margin?”
This guide is for marketplace agencies in Germany, the US and other mature ecommerce markets managing Amazon, Walmart, bol.com, Kaufland, Target, Mirakl retailers, TikTok Shop, Shopify and retail media work for clients with at least a small internal ecommerce team. At that stage, agency software should do more than report performance. It should protect priority.
What competitor advice gets right
The current software landscape has improved a lot. MerchantSpring positions agency analytics around portfolio oversight, scheduled client reports, white-label delivery and connected sales, advertising, profit and operational context. That is useful because many agencies still lose hours rebuilding the same evidence before every client conversation. Their public numbers also make the scale problem clear: agencies may manage hundreds of channels, millions of monthly P&L transactions and recurring reports that cannot depend on manual exports.
Channable and Productsup are strong on feed operations. They explain how product data, channel rules, taxonomy, AI-ready discovery and feed validation help teams launch faster and reduce manual maintenance. That matters for agencies because feed issues are not “just data”. Missing attributes, stale availability and wrong categories create invisible performance losses across Google Merchant Center, marketplaces, social commerce and AI discovery surfaces.
ChannelEngine focuses on marketplace operations: product content, inventory, order management, pricing, promotion, reporting and analytics across a very large channel network. Pacvue is strong on retail media execution. It talks about cross-retailer campaign management, rule-based automation, Buy Box and inventory intelligence, profitability thresholds, share of voice and iROAS. For larger commerce media teams, that is genuinely helpful.
Reddit threads about agency reporting add a more human truth. Agencies are tired of bespoke reporting. Clients still ask for decks even when dashboards exist. Teams stitch Looker, Tableau, PowerPoint, PDFs, spreadsheets and screenshots together because each client wants a slightly different story. The pain is not only data access. It is deciding what deserves explanation, what can wait and what should trigger action before the client asks.
The gap: reporting tells you what changed, not who gets attention first
Most agency software promises some version of visibility, automation and reporting. Good. But a marketplace agency does not only need to know that seven things changed. It needs to know which one should interrupt the team.
That is the gap competitors rarely address directly. A dashboard may show that Client A has a 12% sales dip, Client B has a 38% ROAS improvement, Client C has 91 suppressed SKUs and Client D has a stockout risk. A project-management tool may show 43 open tasks. A reporting tool may schedule all client updates. A retail media platform may suggest bid changes. Still, someone has to decide whether the next hour goes to a Walmart inventory escalation, Amazon negative keyword work, a bol.com price rule, a Mirakl listing fix or a client email.
In a small agency, that decision often lives in the founder’s head. In a 12-person agency, it gets negotiated in stand-ups. In a 35-person agency, it quietly becomes politics: who has the loudest account director, which client pays the biggest retainer, which specialist is available, which issue looks easiest to close before 17:00. None of those are reliable priority systems.
A portfolio risk board gives the agency one shared operating rule: prioritize the combination of client profit at risk, decision window and agency effort required. That is more useful than ranking by revenue, tickets or meeting pressure.
How to score portfolio risk
Keep the scoring simple enough that account teams will actually use it. I like a 100-point model with four components.
1. Profit exposure: 40 points
Profit exposure asks how much money is likely to be lost, wasted or delayed if the agency does nothing. It should use contribution margin, not gross sales. A €30,000 revenue issue with 8% contribution margin may be less urgent than a €9,000 revenue issue with 38% contribution margin and a short stock window.
FiveX helps here by connecting marketplace sales, ad spend, fees, returns, stock and product profitability in one place. That means an agency can rank exposure by expected margin movement instead of relying on channel revenue and a nervous emoji in Slack.
2. Decision window: 25 points
Decision window measures how quickly the opportunity or risk expires. Stockouts, promotion ends, Buy Box loss, budget caps, creator posts and marketplace enforcement deadlines decay fast. A monthly reporting concern can usually wait. A top-selling SKU with seven days of stock cannot.
3. Confidence: 20 points
Confidence asks whether the signal is mature enough to act on. A three-hour sales spike after a TikTok video is interesting, but not always a budget decision. A settlement-reconciled margin drop across two weeks is more reliable. A strong board does not treat every alert equally; it labels signals as early, waiting, confirmed or stale.
4. Effort and ownership: 15 points
Effort matters because agencies sell capacity. If a €1,200 margin risk can be prevented with a 20-minute rule change, it deserves a different place in the queue than a €1,500 risk that needs six people, two client approvals and a feed rebuild. Ownership matters too. If nobody owns the next action, the score should increase, not disappear.
Scenario 1: the loud client versus the profitable risk
Imagine a 14-person marketplace agency managing 22 clients. On Monday morning, two issues arrive.
Client NordPeak Outdoor is the agency’s biggest retainer at $18,000 per month. Their ecommerce director wants a revised Amazon QBR deck by Wednesday. The account is healthy: $420,000 monthly marketplace revenue, 18% blended contribution margin, 21% TACOS and no major stock issue. The deck will take six senior hours because the client wants extra competitor screenshots.
Client LumaPet pays a smaller $6,500 retainer. Their Walmart hero SKU sells 46 units per day at $12.40 contribution margin per unit. Stock cover has fallen to nine days because a replenishment shipment is delayed. Walmart Connect is still spending $310 per day on the SKU at a 24% ACOS, and the Amazon equivalent has 31 days of cover with a 32% contribution margin. Moving part of the media push and updating availability rules could protect roughly $3,400 in contribution margin over the next ten days.
Without a portfolio risk board, NordPeak probably wins the morning because the stakeholder is senior and the request is visible. With a portfolio risk board, LumaPet scores higher: profit exposure is concrete, the decision window is short, confidence is high and the fix is small. The agency still serves NordPeak, but it does not spend the only senior hours of the morning polishing slides while a smaller client’s profitable stock position burns.
This is where FiveX’s stock signals, advertising analytics and product profitability view become practical. The board can surface “high-margin SKU, low stock, active spend” as one exception instead of asking a specialist to notice it across three systems.
Scenario 2: the beautiful ROAS trap
Now take a German homeware client, Haus & Herd, selling on Amazon.de, Kaufland and Otto. The Amazon Ads dashboard shows a lovely improvement: Sponsored Products ROAS rose from 4.1 to 5.6 in eight days. The client asks the agency to increase budget by 25% before the weekend.
The portfolio risk board disagrees. FiveX shows that the winning campaign is concentrated on two ceramic pan sets. One has a selling price of €39.95, marketplace and fulfilment fees of €9.80, landed cost of €16.40 and an expected return and damage reserve of €3.20. At the current CPC and conversion rate, the campaign leaves about €2.10 contribution margin per order after ads. The second pan set looks better in ROAS, but its return rate has moved from 6% to 14% after a packaging change. The return signal is still maturing.
The board scores this as high confidence for “do not scale yet”, medium profit exposure and a short decision window because the weekend budget request is live. The recommended action is not “increase budget 25%”. It is: hold the current budget, move €600 of test spend to the lower-return pan set, set a return-rate review trigger for Wednesday and ask the client to approve a packaging investigation before the next scale decision.
That is operator work. It is less glamorous than celebrating ROAS, but it prevents the agency from scaling an incomplete margin story.
Scenario 3: the feed issue that looked like a content task
A US agency manages BrightDesk Supply, a B2B office accessories brand selling through Amazon Business, Walmart Marketplace and a Mirakl-powered retailer. A feed validation report shows 184 products missing one attribute required by the Mirakl retailer’s new category structure. The task looks boring. Someone labels it “content cleanup”.
The portfolio risk board changes the conversation. Those 184 products include 27 SKUs that generated $52,000 in the last 30 days at a 29% contribution margin. If the missing attribute suppresses them for even five days, the expected contribution margin at risk is roughly $2,500, before any ranking damage. The fix requires one feed specialist for 90 minutes because the attribute can be mapped from an existing ERP field.
High exposure, short effort, clear owner. That task should beat most reporting cosmetics. Productsup-style feed logic is right that scalable feed operations protect revenue; the missing layer is portfolio priority. Which feed issue matters first? The answer should be based on SKU profit exposure and suppression risk, not the order in which tickets were created.
What the board should show every morning
A useful portfolio risk board is not a giant BI project. Start with seven columns:
- Client and channel: for example, LumaPet — Walmart, Haus & Herd — Amazon.de, BrightDesk — Mirakl.
- Commercial object: SKU, campaign, listing group, stock shipment, promotion or reporting decision.
- Risk type: margin, stock, ads, Buy Box, content, returns, payout, client approval or scope.
- Profit exposure: expected contribution margin at risk or upside available.
- Decision window: today, 48 hours, this week or monitor.
- Recommended next action: pause, shift budget, escalate, fix feed, update price rule, wait for evidence or ask client for approval.
- Owner and capacity cost: who acts and how much agency time it consumes.
The board should be reviewed before the daily stand-up, not after. Otherwise the stand-up becomes a theatre of whatever people remembered to mention.
Where agencies usually overcomplicate it
The first overcomplication is trying to score everything. Do not. Start with the top 20 exceptions across the portfolio. If an exception does not affect margin, stock, spend, listing availability, client trust or agency scope, it probably does not belong on the morning board.
The second overcomplication is building the board only for leadership. Leadership needs the portfolio view, yes, but specialists need the same priority logic. A PPC specialist should know when not to touch a campaign because the stock signal is red. A feed specialist should know which attribute issue deserves priority because it protects €2,500 of contribution margin. An account manager should know when a client approval delay has become commercial exposure.
The third overcomplication is confusing automation with autonomy. A risk board can recommend actions and trigger rules, but not every high-risk item should auto-change. Some actions need approval because they affect pricing, channel strategy or client cash flow. The point is not to let software run the agency. The point is to make sure humans spend judgment where judgment is worth the most.
How FiveX fits into the workflow
FiveX is useful for this because portfolio priority needs connected marketplace evidence. A task manager can hold the action. A BI tool can visualize a number. But the risk score needs sales, ads, profitability, stock, repricing context, product data and channel integrations in the same operating layer.
For agencies, three hooks matter most. First, FiveX gives account teams a cross-client analytics cockpit, so they can compare clients and channels without rebuilding spreadsheets. Second, product profitability connects revenue, fees, returns, landed costs and ad spend, so the board ranks contribution margin exposure rather than noisy GMV. Third, FiveX automation and alerts can route exceptions: low-stock active campaign, Buy Box loss on a high-margin SKU, margin breach after a fee change, or feed/listing risk on a profitable assortment.
The result is not “more dashboards”. The result is a calmer agency. Fewer loud-client priority mistakes. Fewer senior hours spent on low-risk polish. Fewer margin surprises that should have been visible three days earlier. And, importantly, a clearer story for clients: “Here is why we worked on this first, here is the profit we protected, and here is what we deliberately left in monitor mode.”
The operating rule
If your agency manages more than a handful of marketplace clients, attention is now a profit lever. Treat it that way. Do not let inbox pressure, meeting calendars or retainer size silently decide the day.
Build a portfolio risk board. Score profit exposure, decision window, confidence and effort. Review it before work is assigned. Connect it to marketplace data that actually reflects margin, stock, ads and operational status. Then let the board protect both sides of the relationship: better commercial outcomes for clients and healthier delivery margins for the agency.
That is the quiet advantage of good agency software. It does not just make reporting faster. It helps the team choose the right work before the wrong work consumes the day.