Share of voice is a seductive metric. It looks competitive, it sounds strategic and it gives everyone in the room a neat percentage to chase. “We own 18% of the sponsored shelf for coffee grinders.” Lovely. Very boardroom. But a marketplace ad account does not become healthier because your logo appears more often. It becomes healthier when the extra visibility converts within the margin, stock and cash limits of the SKU.
The named mistake I see in managed marketplace advertising is buying share of voice like it is market share. It is not. Share of voice is rented attention. Market share is demand you can profitably serve. The gap between the two is where many €5K-plus ad accounts quietly lose money.
My stance: Amazon, bol and MediaMarkt share-of-voice targets need profit permission. Not one account-level target. Not “we want 30% SOV on all priority keywords.” Each keyword lane needs a commercial reason to exist: defend, harvest, conquer, launch or learn. If the lane cannot explain its margin, stock cover and conversion threshold, it should not get a bigger bid just because a competitor is visible.
This guide is for brands spending from roughly €5K per month on marketplace ads across Amazon, bol.com and MediaMarkt. At that level, visibility is no longer the hard part. The hard part is deciding which visibility deserves to be bought.
What the market already says about share of voice
The research is useful, but incomplete. SellerApp explains Amazon share of voice as the percentage of visibility your brand captures in search results, ad placements and product pages. Their example is simple: if a wireless charger brand appears in 30 out of 100 relevant searches, it owns 30% share of voice. That is a good entry point because it separates visibility from clicks and sales.
Blue Wheel’s TACoS and ACoS guidance adds the financial layer many SOV articles miss. ACoS tells you ad spend versus attributed ad sales. TACoS shows paid dependency across total revenue. Their strongest point is that “good” efficiency depends on margin and growth goals, not a universal benchmark.
Headline’s Amazon ads management guide is more strategic. It argues against obsessing over low ACoS and frames campaigns like an investment portfolio: exact Sponsored Products as reliable conversion, Sponsored Brands and video as growth, Sponsored Display and DSP as longer-term reach. That portfolio language is useful because SOV is not one thing. Branded defence and competitor conquesting should never be judged by the same visibility target.
SellerMetrics focuses on the management service side: audit waste, stop bleeding, structure campaigns, protect branded keywords, mine profitable search terms and use bid automation. BidX positions automation around keyword, bid and budget optimisation across Amazon and Walmart, with AMC and DSP analytics for deeper measurement. Podean talks more from the global marketplace agency angle: retail media efficiency, budget allocation, strategic alignment and marketplace maturity.
Reddit and seller forums add the messy operator truth. Sellers argue about whether to optimise for ACOS or TACOS, whether high SOV on a launch keyword is worth it, and whether agencies report impressive impression share while profit stays flat. The recurring worry is not “can I get more visibility?” It is “am I paying for attention that would not have made money anyway?”
The missing angle is clear: most content explains how to measure or increase share of voice. Very little explains when a marketplace advertising service should refuse to buy it.
The FiveX angle: share of voice needs permission by lane
Think of share of voice as a shelf-space budget court. Every keyword or placement comes in with a request: “I want more bid, more budget, more impression share.” The operator asks four questions before approving it:
- What role does this visibility play? Defend branded demand, convert high-intent generic demand, steal competitor demand, launch a SKU or learn a market?
- What is the loaded break-even ACOS? Include marketplace fees, fulfilment, COGS, returns, discounts and any recovery costs.
- Can the SKU serve the demand? Check stock cover, Buy Box or offer eligibility, delivery promise and price position.
- What evidence earns the next budget step? Conversion rate, cost per order, TACoS movement, contribution margin or new-to-brand quality.
This is where FiveX fits naturally into an ad management workflow. FiveX connects marketplace analytics, profitability dashboards, inventory insights and advertising performance, so the operator is not managing SOV in the Amazon console, margin in a finance sheet and stock in a separate operations export. Profit permission becomes visible before bids move.
The five share-of-voice lanes
One account-level SOV target is too blunt. Use five lanes instead.
1. Defend: branded and hero-product searches
Defence is where SOV can be worth paying for even when direct incrementality is imperfect. If shoppers search your brand, a competitor ad above your listing can steal cheap demand. But defence still needs a cap. If branded SOV costs €900 per month and only protects €450 contribution margin, the account is not defending; it is overpaying for comfort.
2. Harvest: high-intent generic terms
These are terms like “noise cancelling headphones bluetooth” or “espresso grinder burr”. The shopper is close to purchase, but the click auction is usually expensive. Harvest terms deserve SOV only when SKU margin, content and stock can handle it. Here, break-even ACOS should be strict because the term is expected to convert.
3. Conquer: competitor and substitute terms
Conquesting often looks exciting in a deck and painful in a P&L. Competitor terms can build awareness, but conversion is weaker and CPCs can climb fast. Give conquesting a learning budget, not a blank cheque. If it does not create incremental contribution margin or useful new-to-brand demand, keep it small.
4. Launch: new SKU visibility
Launch SOV is allowed to run above normal efficiency for a short period, but only if the launch has a written payback rule. “We accept 42% ACOS for 14 days to reach 30 reviews and 18 units per day” is a strategy. “We need visibility” is a wish wearing a blazer.
5. Learn: category and placement exploration
Learning spend should buy information. Which category terms convert? Which MediaMarkt placements send qualified traffic? Which bol Sponsored Products queries have enough demand? A learn lane should have a small budget, a short time box and a decision date. If nobody uses the learning, it was just expensive curiosity.
Example 1: DuneGlow sunscreen on Amazon.nl
DuneGlow sells SPF50 face sunscreen on Amazon.nl for €24.95. After referral fees, fulfilment, COGS and a normal return reserve, the SKU has €7.20 contribution margin before ads. That means the clean break-even ACOS is about 28.9%. The brand spends €6,500 per month on Amazon ads and wants to raise SOV on “face sunscreen spf50” from 14% to 28% before a heatwave week.
On paper, the request sounds sensible. Demand is rising, the term is relevant and competitors are visible. But the SKU has 860 units in stock and normally sells 38 units per day. With the heatwave and extra ads, forecast demand is 82 units per day. Stock cover falls below 11 days. The next replenishment arrives in 18 days.
The permission label is not Scale. It is Controlled Harvest. Keep branded defence at 85% SOV because that protects cheap demand. Let the generic term rise from 14% to 20%, not 28%, with a €70 daily cap and a 24% loaded ACOS limit. No broad discovery until inbound stock is confirmed. In FiveX, the operator can see ad spend next to stock cover and SKU margin, so the decision is not “Amazon is hot, push harder.” It is “Amazon is hot, do not burn through the shelf.”
Example 2: KoffieKompas on bol.com
KoffieKompas sells a burr coffee grinder on bol.com NL and BE for €89. The SKU has €21.50 contribution margin before ads when sold via the preferred fulfilment setup. The brand’s bol Sponsored Products budget is €5,200 per month. Current SOV on “koffiemolen bonen” is 11%, and an agency proposes moving to 25% because the term has strong volume.
The hidden issue is marketplace mix. Dutch orders convert at 7.8% and create €15.20 contribution margin after ads at current CPCs. Belgian orders convert at 4.1% because delivery is one day slower and two competitors price €6 lower. In a combined campaign, the blended ACOS looks fine at 19%, but BE clicks are eating the margin.
The right decision is to split the lane. NL gets Harvest permission: raise SOV toward 22% with an ACOS ceiling of 23% and a daily budget of €95. BE gets Learn permission only: €15 per day, exact terms, no competitor expansion and a decision after 200 clicks. FiveX profitability dashboards make the split easier because the operator can compare margin by marketplace and SKU instead of trusting a blended campaign average. Blended SOV is a fog machine. Split SOV is management.
Example 3: VoltEdge router on MediaMarkt
VoltEdge sells a Wi-Fi 7 router through MediaMarkt marketplace for €179. The product has €32 contribution margin before ads and a practical break-even ACOS of 17.9% after expected returns. The MediaMarkt retail media campaign shows a 5.4 ROAS on category placements, and the team wants 35% SOV on “wifi 7 router”.
Then price position changes. A larger competitor drops to €159 for ten days. VoltEdge can match the price, but margin would fall from €32 to €14. If it keeps price at €179, conversion drops from 3.6% to 2.1%. Either way, the old share-of-voice target is now unsafe.
The permission label becomes Hold for conquest and Controlled Defend for exact product-intent placements. Spend €40 per day to protect shoppers already comparing VoltEdge, but do not chase 35% category SOV while the price gap is active. If FiveX marketplace analytics shows Amazon.nl still converts at full price with 24 days stock cover, move part of the week’s budget there. The goal is not to win every retail media auction. The goal is to put the next euro where it can survive the P&L.
What your agency should report every week
If you outsource marketplace ad management, ask for a share-of-voice report that includes permission, not just percentages. A useful weekly report has seven lines:
- Top 20 SOV keywords or placements by spend and strategic role.
- Current SOV, target SOV and the reason for the target.
- Loaded break-even ACOS per SKU or product family.
- Stock cover after forecast paid demand.
- Contribution margin after ads, not only ROAS or platform sales.
- Budget decision for each lane: defend, harvest, conquer, launch, learn, hold.
- Evidence needed for the next budget increase.
This report should be short. Long SOV decks often hide the actual decision under screenshots. The operator voice is simple: show me what we bought, why we bought it and whether the SKU deserved it.
The operator checklist
- Never set one SOV target for the whole ad account.
- Separate branded defence, generic harvest, competitor conquest, launch and learn budgets.
- Use loaded break-even ACOS, not the old campaign target.
- Check stock cover after expected paid velocity.
- Split NL and BE when conversion, delivery promise or price position differs.
- Reduce SOV targets when price position weakens, even if ROAS looked good last week.
- Make every SOV increase earn the next step with conversion, margin or TACoS evidence.
How FiveX helps
FiveX helps teams turn share of voice from a vanity target into an operating decision. Marketplace analytics show where Amazon, bol and MediaMarkt demand is moving. Profitability dashboards calculate whether the SKU can afford more visibility after fees, fulfilment, returns and discounts. Inventory insights prevent the classic mistake of winning SOV into a stockout. Advertising automation and AI recommendations then help operators raise, hold or cut bids with the right context.
That is the difference between “we increased visibility” and “we bought the right visibility.” One sounds nice in a report. The other protects profit.
If your brand spends more than €5K per month on marketplace advertising, share-of-voice governance belongs in the ad service cadence. Not because SOV is bad. It is not. SOV is powerful when it has permission. Without permission, it is just an expensive way to be seen losing money.