What to Actually Demand from an Amazon PPC Agency (Account Load, DSP, Creative, and Attribution)

Quick answer: Most Amazon PPC agencies run 12–15 accounts per specialist — enough to monitor campaigns, not enough to build them. Demand a hard account-load number, DSP included in standard scope, a weekly creative cadence, and attribution that goes beyond last-click Sponsored data. Those four requirements separate reactive maintenance from a program that actually compounds.

Most Amazon PPC agencies are running 12–15 accounts per specialist. At that load, your account gets checked twice a week if you're lucky. The bids get touched. The obvious waste gets cut. And the strategic work — DSP sequencing, creative testing cadence, attribution logic — never happens, because there's no time. The agency looks busy. Your account looks managed. And your growth plateaus in a way that's easy to blame on the category.

This is the pattern that's been repeating across the industry for years. As of 2026, it's still the norm.

The Account Load Problem No One Talks About in Their Pitch Deck

The math is simple and damning. A specialist managing 15 accounts in a 40-hour week has roughly 2–3 hours per account per week. That's enough for reactive maintenance — catching bid drift, pulling search term reports, flagging a suppressed listing. It is not enough for proactive strategy.

What gets cut first when load is high isn't the work that shows up on a dashboard. It's the work that requires reading the account: DSP audience builds, creative brief reviews, search term mining below the top 20 queries, bid adjustments on placement-level data, and any analysis that requires understanding why something is happening, not just that it is. That's the work that compounds. And it's the first thing that disappears when a specialist is stretched.

The tell in an agency pitch is simple: ask the account lead to name your top five SKUs by unit velocity and your current TACoS trend. If they can't answer without looking it up, they're not running your account — they're monitoring it.

Joe Singe founded LSD after personally managing 23 active accounts at a previous Amazon agency and watching this pattern repeat without exception. Overloaded specialists. Account leads who couldn't name half their clients' top SKUs. Brands paying enterprise rates for part-time attention. The 6-account cap per specialist is the direct structural response to that experience — not a marketing line, but the constraint the entire service model is built around.

What the cap actually enables: bi-weekly strategic reviews with a prepared account lead, same-day response on most issues, and 20-minute response on critical issues — ad disapprovals, listing suppressions, Buy Box loss on a top ASIN. That last one matters more than most brands realize until it doesn't happen.

Sponsored Campaigns Are Not a Strategy — They're a Starting Point

Sponsored Products, Sponsored Brands, and Sponsored Display are intent-capture tools. They intercept shoppers who are already in the market, already searching. That's necessary. For a mature CPG brand trying to build category share, it is not sufficient.

Most Amazon advertising agencies treat Sponsored as the full scope because it's the easiest to report on. CPC, ACOS, ROAS — clean metrics, clean slides. The harder work (DSP, creative iteration, attribution architecture) is harder to attribute and harder to defend in a monthly deck. So it doesn't happen.

A real Amazon advertising program sequences differently. The Laser Focused Blueprint frames it explicitly: SEO drives organic traffic → CRO converts that traffic → Sponsored PPC scales what's already working → DSP extends reach and re-engages audiences outside the intent window. Each step builds on the last. Running Sponsored without CRO is spending to amplify a broken page. Running Sponsored without DSP leaves the mid-funnel and the off-Amazon audience entirely unaddressed.

Ask any prospective Amazon PPC agency: what percentage of your Amazon clients are running DSP, and is it included in scope or billed separately? The answer tells you exactly how they think about the channel.

DSP Should Be Standard Scope, Not an Upsell

Amazon DSP is a programmatic display channel — CPM-priced, impression-based, reaching audiences across Amazon-owned properties and third-party inventory using Amazon's first-party purchase data for targeting. It is not Sponsored Ads with a bigger budget. It reaches people who haven't searched yet, or who searched and left without buying.

The most common agency model: DSP is a separate managed service, separate retainer, often a separate team. The result is a Sponsored team and a DSP team that don't talk to each other, running disconnected strategies against the same customer. Sponsored search term data never informs DSP audience segments. DSP retargeting data never informs which Sponsored placements are worth defending. The two channels coexist without coordinating.

LSD includes Amazon DSP as part of standard PPC scope — not an add-on — because the two channels are dependent, not parallel.

| Sponsored Ads | Amazon DSP | |---|---| | Intent-capture: intercepts active searchers | Demand-building: reaches audiences before or after search | | CPC pricing — pay per click | CPM pricing — pay per impression | | Amazon search results and detail pages | Amazon-owned properties + third-party inventory | | Short attribution window (7-day click default) | View-through attribution across longer consideration windows | | Best for: in-market shoppers ready to buy | Best for: retargeting, conquesting, lapsed buyer re-engagement | | Easiest to report on; most agencies stop here | Harder to attribute; most agencies upsell it separately |

For CPG specifically, DSP earns its budget in three scenarios: retargeting high-intent audiences who viewed a detail page but didn't convert, conquesting competitor ASINs with in-market audiences, and reaching lapsed buyers in replenishment categories before they default to a competitor brand.

DSP does not earn its budget before listings convert at a competitive rate, before Sponsored campaigns have found their footing, or when the retargeting window is so wide the audience has long since made a purchase decision elsewhere. There is a minimum managed spend threshold through a certified DSP partner — this is not a channel to test with a rounding-error budget.

Creative Is Where Most Amazon Agencies Have No Answer

Listing images, A+ content, Sponsored Brand creatives, DSP display units, and video assets all decay. What converts in Q1 may be stale by Q3, especially in competitive CPG categories where competitors are iterating on their own creative every few weeks.

The typical agency response: creative is out of scope, or the brand handles it, or there was a one-time listing build at onboarding and nothing since. That's not a creative program. That's a queue.

Across our enterprise client portfolio, the pattern we see repeat is that creative fatigue is invisible until it's expensive. Conversion rate drifts down slowly enough that no single week looks alarming. By the time the problem is obvious, a competitor has moved in on the category terms that used to be easy wins.

A real creative cadence looks like this: weekly asset production, structured A/B testing on hero images and A+ modules, Sponsored Brand video refreshed on a seasonal and promotional calendar, and DSP display creative matched to the audience segment being targeted. Not requests fulfilled — a program running.

The Sightline AI Engine runs on a Mon-to-Fri weekly cadence: brief Monday, generate Tuesday, review Wednesday, finalize Thursday, ship Friday. It produces static and video assets across every channel — hero images, lifestyle shots, infographics, A+ and Premium A+ modules, Sponsored Brand creatives, DSP display units, and TikTok Shop video. Brand guardrails (typography, palette, voice, talent archetypes, do-not-ship list) are layered at kickoff. Brand environments stay isolated per client — no cross-client training, no bleed. The brand owns every asset outright.

The practical question to ask any Amazon PPC agency: who produces your creative, on what cadence, and how do you test it? If the answer is "the brand provides assets" or "we have a design team that turns around requests," that is a queue, not a program.

Attribution on Amazon Is Broken by Default — Here's How to Fix It

Amazon's default attribution window is 7 days for Sponsored Products and 14 days for Sponsored Brands. A sale that happens 8 days after a Sponsored Products click gets credited to organic, not to the ad that drove the consideration. That's not a quirk — it's a systematic undercount of paid contribution.

Most agencies report on Amazon's native attribution data because it's easy to pull and easy to present. The problem: it systematically undercounts DSP contribution, overcounts last-click Sponsored, and tells you nothing about what's happening off Amazon. For brands running Google Ads to Amazon via Attribution tags, the gap is even larger — clicks that drive Amazon purchases often show as zero-revenue in Google's reporting unless Attribution is configured correctly.

When we audit new accounts, the most common gap isn't bid strategy. It's attribution logic. Brands are making budget allocation decisions based on last-click Sponsored data that's missing a meaningful share of the actual conversion story — and the channels that look "inefficient" are often the ones doing the upstream work.

A mature attribution stack for an enterprise Amazon brand includes:

• [Amazon Attribution](/blog/what-is-amazon-attribution-for-advertising) for tracking off-Amazon traffic sources (Google, Meta, email) driving Amazon purchases
• Amazon Marketing Cloud (AMC) for cross-channel path analysis and overlap reporting
• A third-party tool — Triple Whale, Northbeam, or GA4 with custom event tracking — for DTC-to-Amazon overlap and blended ROAS visibility

The agency's job is not to hand you a ROAS number. It's to tell you which part of the funnel produced it and whether the mix is defensible. If your Amazon PPC management team can't explain the difference between attributed sales and total sales lift, the reporting is decorative.

The Reporting Trap: Vanity Metrics vs. Decisions

ACOS and ROAS are the two metrics most Amazon agencies lead with in monthly reviews. Both are useful. Neither tells you whether your account is growing or efficiently spending against a shrinking opportunity.

The metrics that actually indicate account health:

• [Organic rank trajectory](/blog/amazon-seo-2025) on primary category keywords
• Conversion rate by ASIN and by traffic source
• New-to-brand (NTB) percentage on Sponsored campaigns
• Share of voice on top category terms
• DSP view-through contribution to total attributed sales

New-to-brand rate deserves particular attention for CPG brands. If the majority of your Sponsored sales are going to repeat buyers, you're subsidizing purchases that would have happened anyway. A healthy program grows NTB share over time. If your agency isn't tracking it, they're not measuring whether your advertising is actually building the business.

Bi-weekly strategic reviews — not monthly slide decks — are the right cadence for mature accounts. Monthly reporting is too slow to catch bid drift, creative fatigue, or a competitor's promotional push before it erodes your position.

What to demand from every reporting format: a clear decision attached to every major metric. Not "ACOS was 18% this month" but "ACOS was 18%, down from 21%, driven by negative keyword additions on these three campaigns — next action is to test a lower bid on top-of-search placements where ROAS is below threshold." The difference between reporting and analysis is whether it tells you what to do next.

Catalog and Listing Operations: The Work That Never Shows Up in the Pitch

PPC performance is downstream of listing quality. A Sponsored Products campaign driving traffic to a listing with a weak hero image, missing bullet points, or a suppressed Buy Box is amplifying a broken page. That's the Laser Focused Blueprint's core warning, and it applies whether you're spending $10,000 a month or $1 million.

Catalog ops work that most agencies don't include in scope:

• ASIN suppression recovery
• Variant consolidation and parent-child restructuring
• Compliance flag resolution
• Vendor Central cost negotiation support
• Inventory signal monitoring to prevent stockouts from tanking organic rank

Brand protection belongs in scope too. A hijacked listing running on your Sponsored traffic is a direct conversion leak — shoppers click your ad, land on a listing controlled by someone else, and buy a counterfeit or a competitor's product. Hijacker removal, IP filings, and counterfeit test buys are not glamorous. They are load-bearing.

International catalog management compounds every one of these issues. A listing optimized for US search behavior does not port cleanly to UK, EU, or Japan. Different keyword structures, different compliance requirements, different shopper intent patterns. In Q1 2026, we're seeing more enterprise CPG brands expand to EU and Japan simultaneously — and the brands that treat it as a copy-paste job from the US catalog pay for that assumption in suppressed rank and conversion rate.

The practical test: ask a prospective agency what happens when one of your ASINs gets suppressed on a Friday afternoon. Who notices, how fast, and what's the recovery process? If the answer is vague, catalog ops aren't in scope.

How to Actually Evaluate an Amazon PPC Agency Before You Sign

The pitch deck won't tell you what you need to know. The questions will.

Ask for the account load number — not the agency average, the specific specialist who would run your account. If they won't give it, assume it's high.

Ask whether DSP is in scope or a separate engagement. If it's separate, ask how the Sponsored and DSP teams coordinate. No clear answer means the channels won't be coordinated.

Ask to see a real creative testing log from an existing client — not a case study, an actual record of what was tested, when, and what the result was. This surfaces whether creative iteration is real or aspirational.

Ask how attribution is set up for a brand running both Amazon and DTC. "We use Amazon's native reporting" is an incomplete answer by design.

Ask what the response SLA is for a critical issue — ad account suspension, listing suppression, Buy Box loss on a top ASIN. The answer should be specific and should come with a process, not a promise. LSD's critical-issue response time is 20 minutes, with a defined escalation path behind it.

In our experience managing $450M+ in client ad spend across 50+ enterprise brands, the agencies that can't answer these questions specifically are the ones where your account gets managed reactively — checked when something breaks, not built when everything is working.

What a Real Amazon PPC Program Looks Like at Enterprise Scale

The right Amazon advertising agency is not the one with the most clients or the most impressive logos on the homepage. It's the one where a senior specialist can walk into a call and tell you your top five ASINs by velocity, your current organic rank on your three primary category terms, and which Sponsored campaign is pulling the most new-to-brand volume — without looking it up.

At enterprise scale, the constraint is not budget. It's attention.

The brands that compound on Amazon are the ones whose agency is running proactive strategy 40 hours a week on their account, not reactive maintenance 3 hours a week. The difference between those two modes is almost entirely a function of account load.

The 6-account cap is not a boutique quirk. It's the structural answer to the attention problem. It's why LSD carries 98% client retention across 50+ brands — not because every campaign is perfect, but because every problem gets caught before it compounds and every opportunity gets acted on before a competitor does.

If your current agency can't tell you your NTB rate, your DSP view-through contribution, and your organic rank trajectory in the same conversation — that's the audit. You already know what's missing.

LSD offers a 48-hour audit for any Amazon channel. No pitch, no pressure. The brand keeps the audit either way. If the gaps are there, the audit will surface them.

Frequently Asked Questions

How do we evaluate whether our current Amazon PPC agency is actually managing our account or just monitoring it?

Ask your account lead — without prep time — to name your top five SKUs by unit velocity, your current TACoS trend, and the last structural change they made to your DSP audience builds. If they need to pull a report to answer the first two, they are monitoring your account, not managing it. A specialist running six accounts can answer those questions cold; one running fifteen cannot.

At what revenue threshold does Amazon DSP become worth running alongside Sponsored campaigns?

DSP is less about revenue threshold and more about catalog maturity and audience size — specifically, whether you have enough purchase and browse data for Amazon's targeting to be meaningful. In practice, brands with established category presence and a mid-to-high five-figure monthly Sponsored budget typically have enough signal to make DSP retargeting and conquest audiences work. Running DSP before that point often means paying CPM for audiences too thin to optimize against.

Our internal team handles Sponsored Products well. Is there still a case for bringing in an agency just for DSP and attribution?

Yes, but only if the agency runs both channels under one roof — otherwise you recreate the coordination problem the post describes, just with your internal team on one side and an external DSP team on the other. The value of integrated DSP management is that Sponsored search term data directly informs DSP audience segments and vice versa; that loop breaks the moment the two channels are owned by separate teams who report separately.

How often should creative assets actually rotate on Amazon, and what does a realistic production cadence look like for an enterprise CPG brand?

For active Sponsored Brands and DSP placements, creative should rotate on at least a monthly testing cadence — stale assets lose click-through rate as the same audiences see them repeatedly, and Amazon's relevance signals reward fresh content. At LSD, the Sightline AI Engine runs a weekly production cycle (brief Monday, generate Tuesday, review Wednesday, finalize Thursday, ship Friday), which means enterprise brands can test new hero images, A+ modules, and DSP display units continuously without the bottleneck of a traditional creative agency retainer.

Which attribution model should we use to evaluate Amazon PPC performance — Amazon's native attribution, a third-party MTA tool, or last-click?

Amazon's native attribution defaults to a 14-day click, 14-day view window, which inflates DSP's apparent contribution and can make Sponsored Display look stronger than it is against tighter measurement windows. For CPG brands running both Sponsored and DSP, the most defensible approach is to run Amazon's native reporting alongside a vendor-agnostic tool — Triple Whale, Northbeam, or GA4 with proper UTM architecture — and reconcile the gap rather than trusting either source alone. Last-click attribution on Amazon systematically undervalues upper-funnel DSP work and should not be the primary decision-making metric.

If we switch Amazon PPC agencies, how long before we should expect to see a meaningful impact on performance — and what should happen in the first 30 days?

Expect a 60–90 day ramp before performance changes are clearly attributable to the new agency's strategic decisions, because Amazon's algorithm takes time to respond to structural changes in campaign architecture, bid strategy, and listing quality. In the first 30 days, a competent agency should complete a full account audit, identify the highest-priority gaps in campaign structure and listing CRO, and begin executing — not still be in onboarding. If week four is still discovery calls and slide decks, the timeline is already slipping.