Every week, a marketing director asks their agency the same question: “Should I use Automatic or Manual targeting on Amazon?”
It sounds reasonable. It sounds strategic. But it’s actually the wrong question entirely.
Think of it like asking a chef whether they prefer a knife or a stove. The answer isn’t one or the other. The answer depends on what you’re cooking and what the ingredients are telling you.
I’ve spent years inside Amazon advertising, managing millions in ad spend for clients who care about real growth-not just vanity metrics. And I’ve learned something that most advertisers miss: the Auto vs. Manual debate is a distraction. It frames your job as choosing between two tools when your real job is understanding what those tools reveal about your business.
Let me show you what I mean.
The Old Way of Thinking Is Broken
Here’s the conventional wisdom you’ll find in any Amazon advertising guide:
- Automatic targeting: Amazon’s algorithm finds keywords for you based on your listing. Good for discovery. Great for new products.
- Manual targeting: You pick the exact keywords. You control match types and bids. Good for scaling winners.
None of that is wrong, exactly. But it’s shallow. It treats Amazon advertising like a light switch-flip it one way for discovery, the other way for control.
Real growth doesn’t work that way.
At our agency, we built everything around a different principle: alignment with client outcomes. We don’t just run campaigns. We build systems that generate intelligence. And when you view Automatic and Manual through that lens, the entire conversation changes.
Automatic Targeting Is Not a Campaign. It’s a Laboratory.
Most advertisers launch an Auto campaign and get frustrated within two weeks. “Why am I showing up for red sneakers when I sell blue boots?” They blame the algorithm. They pause the campaign. They retreat to Manual.
This is a mistake.
Automatic targeting is not a spending vehicle. It is a research laboratory.
Here’s what’s actually happening under the hood. When you run an Auto campaign, you’re pinging the Amazon A9 algorithm. You’re asking it: “Based on my listing, my title, my bullet points, my A+ content-who do you think my customer is?”
Amazon answers by serving your ad to relevant searches. Those searches are not random. They are Amazon’s best guess at your product’s identity. And that guess reveals something important about your business.
There are three possible outcomes, and each one tells you something different:
Outcome 1: Your Auto campaign generates mostly high-converting, relevant traffic.
This means Amazon understands your product. Your listing is tight. Your positioning is clear. Your customer match is strong. The system is working. You have validation. Now take those winning terms and move them into Manual.
Outcome 2: Your Auto campaign generates a mix of good and irrelevant traffic.
This is normal. Amazon is testing the edges of your product’s identity. Some of that “irrelevant” traffic is actually Amazon exploring adjacent customer needs. This is valuable data for expansion. Don’t panic. Just monitor and learn.
Outcome 3: Your Auto campaign generates mostly garbage.
This is the one nobody wants to talk about. But here’s the truth: if your Auto campaign is full of junk search terms, the problem is not Amazon. The problem is your listing.
Amazon is confused because your product page is confusing. Your title might be too generic. Your bullet points might lack specificity. Your category node might be wrong. The algorithm cannot read your mind. It reads your content.
The fix is not to turn off Auto. The fix is to rewrite your listing until Amazon understands exactly who you serve and why they should care.
Manual Targeting Is a Factory, Not a Control Room
The conventional wisdom says Manual targeting gives you “control.” You choose keywords. You set match types. You decide bids.
But control over settings is not the same as control over outcomes.
Manual campaigns are operations. They are the factory floor where you scale what the lab discovered.
Here is the insight that changes everything: Manual campaigns are dead without a data moat.
Most advertisers take their top 20 search terms from an Auto campaign, dump them into a Manual Exact campaign, and call it strategy. It’s not. It’s copy-paste.
Real strategy requires understanding the character of each keyword. Not just the volume. Not just the ACOS. But the context of the customer intention behind the search.
Here’s how we think about it:
- Low ACOS + High Volume: This is a price-taker keyword. You own the category. Your Manual Exact campaign is just collecting tax. Maintain position. Protect margin.
- High ACOS + High Volume: This is a discovery keyword. Customers are searching broadly. They don’t know your brand yet. This belongs in a Manual Broad or Phrase campaign as top-of-funnel traffic. Don’t kill it. Nurture it.
- Low Volume + High Conversion: This is your long-tail moat. These are specific customers who know exactly what they want. Bid aggressively here. This is your profit center.
The “control” of Manual is useless without the context of the customer. And that context comes from understanding your data-not just looking at reports, but living inside them.
This is why we limit the number of clients per digital marketing manager. You cannot build deep keyword intuition when you are managing 30 campaigns for 10 different brands. You need focus. You need to breathe the data.
The Real Framework: The Gap Analysis
Here is a tactical approach you rarely see discussed anywhere. We call it the Gap Analysis, and it changes how you allocate budget and attention.
Step 1: Run a high-budget Auto campaign for 90 days.
Do not touch the negatives. Let it run wide. Let Amazon explore every corner of your product’s semantic neighborhood. This is your discovery phase. You are not optimizing for efficiency yet. You are optimizing for intelligence.
Step 2: Compare your Auto search terms against your Manual campaigns.
Look at the top 50 converting search terms from your Auto campaign. Compare them against your active Manual campaigns. The gap between these two lists tells you everything you need to know.
Here’s what to look for:
- If 80% or more of your Auto terms already exist in your Manual campaigns: Your listing is perfectly optimized. Your keyword mapping is complete. Amazon and you agree on who your customer is. Your growth lever now shifts to creative innovation-video, lifestyle imagery, better A+ content that tells a story.
- If 30% or more of your Auto terms are new, high-converting terms not in your Manual campaigns: You have a data lag. Your Manual campaigns are stale. You are running on old intelligence while the algorithm has moved forward. This is the most common failure mode. The fix is aggressive negative keyword management in Auto and rapid expansion of Manual campaigns with new winners.
- If your Auto terms are mostly junk: Pause spending and fix your listing. Every dollar you spend on ads for a poorly optimized listing is a dollar wasted. The algorithm cannot sell what it cannot understand. Invest in better copy, better images, better category placement. Then relaunch.
This framework transforms Auto and Manual from competing choices into a single intelligence loop.
Automatic discovers what Amazon believes about your product.
Manual validates what the market will pay for that belief.
They are not opponents. They are partners in a continuous cycle of learning and scaling.
What This Means for Your Business
If you are focused on long-term growth-not just next month’s numbers-this changes how you evaluate your Amazon advertising.
Stop asking your team: “Are we running Auto or Manual?”
Start asking:
- What is our data telling us about our product’s identity?
- Are we building intelligence, or are we just spending budget?
- Do we have the right systems to turn advertising data into strategic decisions?
The brands that win on Amazon are not the ones with the best bids. They are the ones with the best understanding of their customer and the discipline to let the data guide their strategy.
We built our agency around this idea. We limit our client load so we can focus. We use Slack to stay in constant communication. We build custom BI dashboards so every decision is grounded in real data-not gut feelings. And we structure our agreements around outcomes, not hours.
The Auto vs. Manual debate will rage on in forums and YouTube comments forever. But for serious advertisers who want real results, the question is not which targeting mode to use.
The question is: What are you trying to learn?
Answer that honestly, and the right campaign structure becomes obvious.
