How to find & Leverage Negative Keywords in Amazon PPC Advertising Campaigns

In this article we provide you with all information you need to identify negative keywords for your Amazon PPC campaigns. Based on our observations, amazon negative keywords are one of the most under-appreciated aspects of Amazon PPC.

What are Amazon Negative Keywords?

To begin, let’s define what negative keywords for Amazon PPC are?

Negative keywords allow you to block a particular search term from triggering your Amazon Ads when someone searching on Amazon uses that particular term as part of their search query. Think of negative keywords as a black list of keywords in your Amazon PPC campaigns.

Assume you are and Amazon FBA business selling chef knives and you are bidding on the term “chef knife.” Your ad can be placed into the Amazon PPC auction and be shown to the shoppers if the keyword “chef knife” triggers the match type.

Even if a buyer searches for a “German chef knife,” your Japanese chef knife may come up. Because your keyword phrase “Chief knife” appears in their search.

In this case, it may be a good idea to include “German” or “Germany” as a negative keyword so that any consumer searching for “German” isn’t matched with your Japanese chef knives.

We add “German” as a negative because consumers expressly seeking German knives are less likely to buy our Japanese-made blades, and well often have statistics (Poor ACoS, Low CVR) to back up this claim before adding the negatives.

The Easiest Ways to Quickly Find Negative Keywords for Amazon

While we will discuss some advanced tactics to identify negative keywords to add to your Amazon ad campaigns, there are also a few very obvious signs that you should definitely negative out a certain search query. These are:

• Keywords with many clicks but few or no sales conversions (aka low conversion rate)
• Keywords a high cost per sales (or cost per action)
• Keywords with very high CPCs (significantly above your account average) and no conversions
• Keywords with a significant amount of impressions, but very few clicks (i.e. a low CTR)

Why is it Important to add Negative Keywords to Amazon Advertising Campaigns?

Not adding negative keywords is a pretty frequent problem in many Amazon PPC accounts. Many Amazon Sellers simply ignore it and continue to waste their ad spend. Adding negative keywords is often the lowest hanging fruit for improving ad account performance.

Adding negative keywords is a fantastic way to keep your advertising from being seen by those who aren’t directly interested in what you’re selling. Remember, the goal is to pay for clicks to customers who are most likely to purchase your product.

Hence the main and most important reason to use negative keywords is to reduce your advertising costs by avoiding wasted ad spend on irrelevant search terms that could otherwise be spent on other keywords with higher relevancy and a better performance.

Different Amazon Keyword Match Types and How they Impact Negative Keyword Targeting

Before we go into the specifics of Amazon negative keywords, it’s vital to have a firm grasp on Amazon PPC’s three main keyword match types (Exact, Phrase, and Broad). If you are unfamiliar with this concept, check out our article on different Amazon PPC keyword match types.

As a basic primer, refer to the table below. The term “baby towel” will be used as an example to demonstrate how different Amazon keyword match types work. We’ll go through the various keyword match types available: exact, phrase, and broad.

Overall, while choosing keywords and match types, keep in mind that as you move from exact match, to phrase match and broad match, you cast a wider net and reach a larger audience. While this gives you the opportunity to show your Amazon product listings to more potential shoppers, you also forfeit targeting precision and your chances of getting the perfect consumer with every ad impression decrease.

Does that make sense? Now that we’ve covered match types, let’s return to negative keywords and how they fit into the picture.

Exact Match Negative and Phrase Match Negative Keywords

Negative keywords are used to filter out search queries. They prevent your ads from showing when a shopper uses a “negative keyword” as part of their search query on Amazon.

Negative Exact Match Keywords

A negative exact match type will hide an ad from shoppers who are looking for that exact term or near misspellings and plurals.

Negative Exact Match Keywords

A negative phrase keyword, on the other hand, will prevent ads from showing for any search queries that include the negative phrase match keyword. There are no negative broad match types.

Different Categories of Negative Keywords for Amazon FBA Businesses

There are different types of negative keywords and in some cases unearthing them is straightforward, in other cases things can be tricky. Let’s dive right in:

• Competitors’ Brand Keywords
  While a case can often be made for purposely bidding on competitors’ brand names, in many cases you may also want to avoid doing that. In many cases shoppers who are already searching for a competing brand have already made up their mind and serving your ads to them may not be cost efficient. In practice, you may setup a specific ad campaign to target competitors’ keywords and thus want to exclude them from all your other campaigns – as negative keywords.
  
• Unfit Product Features & Characteristics
  We are back to the “Japanese knife set” example. Features or attributes that clearly do not fit your Amazon product listings should routinely be added as negative keywords. Think of factors, such as materials, origin, color or product sub-category. Basically you want to brainstorm all potential keywords that shoppers could think of in the context of your product that do not all apply to your products.
  However, sometimes intuition may be misleading. For example, what if most people who were looking to buy knife sets on Amazon bought them for the purpose of gifting and really didn’t care about whether they get Japanese of German knifes? While in the case of knifes this is most likely not the case, this may apply to other product types. It is thus important to understand the context and intent of shoppers who buy your products and to only include terms that will definitely not trigger a sale!
  
• Search Terms that are – Statistically Speaking – not working
  So this category is tricky. What if you have spent a certain amount of budget on a set of keywords without seeing any sales coming in. Should you exclude them and add them as negative keywords? What if the next ad click on such a keyword actually triggered a sale? This is were you need to understand the concept of statistical significance. The accurate definition of statistical significance is:
  
  “the probability of the null hypothesis being true compared to the acceptable level of uncertainty regarding the true answer”. Or to put it simple: statistical significance measures the likelihood that a certain event (in this case not generating a sale) is caused by chance (bad luck!) or can be explained by an underlying factor (i.e. your product listing is not a good fit for a given search term).
  
  So this is where things get complicated and some number crunching is required to make statistically accurate decisions that actually help you maximize Amazon FBA sales.
  

Which Data to Analyze to Spot Negative Keywords in your Amazon Advertising Account?

As outlined in the past section, in many cases identifying irrelevant search terms is straightforward and deductive reasoning is sufficient to add certain keywords to your “black list”. But how about search terms where things are not as clear cut? How do you figure out which search terms are to be added as negative keywords? Is there a common threshold to trigger a negative keyword? What is considered a good or a bad search term?

In practice, evaluating whether or not a certain keyword will bring in the right kind of shoppers to your listing can be difficult. For example, should the keyword “kitchen knife” be included as a negative keyword if you are selling the chef knife? Sometimes the decision can be made be looking at the data.

Analyzing the Amazon Search Term Report

A great starting point is analyzing historic Amazon ad performance. The data we’re talking about comes in the form of a Search Term Report in your Advertising Console account. This displays what consumer queries are being shown as an impression for your Amazon PPC campaigns.

To download your Search Report. Go to the side nav and choose “Report” in your Amazon ad console.

Once in the Report Screen, click “Create report”

Select the report type “Search Term”, click “Run report”. You create a search term report for the following ad types:

• Sponsored Products
• Sponsored Brands
• Sponsored Brand Videos

Download the report, and examine which search terms that are not reaching your performance objectives or poor KPIs (ACoS, CVR, CTR) by using filters in Excel or any other spreadsheet program. These search terms with poor metrics will be the candidates to be added as negative keywords.

How to Use Statistical Confidence Analysis to Decide which Keywords to add as Negative Keywords?

Now that we know how to find and export all the search terms our Amazon ads have triggered we need to decide which ones to add as negative keywords – based on their performance.

Let’s take an example: Say keyword A has triggered 50 clicks and had zero conversions, but keyword B which has also accumulated exactly 50 clicks had one sales conversion. Is keyword A “working” and keyword B is “not working” or are differences in performance coincidental?

This is where a statistical significance test comes in handy. Without going into too much detail, when it comes down to deciding whether or not A has outperformed B, you want to conduct what is called a two-sided significance test. The good news is that you do not need a PhD in statistics to check whether the performance of two keywords is really different, or if differences in performance are caused by chance. You can use any statistical significance calculator that is available online (for example here).

Simply input the number of ad clicks in the “visitors” box, input the number of sales conversions and then select “two-sided” and select a 90% or 95% confidence interval. 95% is better, but you may need many clicks to get results, so 90% is a good fallback option.

The result of the test will show you whether keyword B really performed better than keyword A, or if it might be better to wait a bit longer.

If you’re utilizing amazon PPC management software like SellerMetrics to assist with adding negative keywords at scale, you can simply define a metric-based rule to suggest negative keyword additions, which the software can either apply automatically or present to you for human review. More on that later

Best Practices for Managing Negative Keywords Across Different Amazon Ad Campaigns

Now that you know how to differentiate between performing and under-performing keywords let’s dive into a more tactical of negative keyword utilization. Let’s use chef knives as an example again, and let’s say we’ve gone from selling chef knives to now selling butcher knives as a new product, and we’ve set up two manual campaigns. The first has exact match keywords, whereas the second has broad match keywords.

The problem is that both of these campaigns include phrases that may match our knife products to customers searching for “Japanese steel butcher knives.” In this case, adding “ Japanese steel butcher knives ” as an exact match negative keyword in the broad campaign would be a smart idea.

The idea is to keep a tighter grip on which campaigns are generating ad impressions for which customer queries in order to isolate your account’s search terms. This allows us to have more control over an account’s performance.

If you want to keep your advertising account organized and performing well, you’ll need to have a structured negative keyword strategy.

Advanced Tactics: Mutually Exclusive and Collectively Exhaustive (MECE) Negative Keyword Layering

When using negative keywords to prevent different ad campaigns from triggering for the same keywords (and potentially competing against each other) is important to implement a Mutually Exclusive and Collectively Exhaustive (MECE) negative keyword targeting strategy.

The image below shows what we mean by that. Let’s say we target “baby towel” as an Exact Match (EM) keyword in once campaign and as a Phrase Match Keyword (PM) in another Amazon ad campaign. In this scenario our keyword targeting is not exclusive and the campaigns are competing for the same clicks.

We can fix this by adding “Baby Towel” as a negative exact match keyword to one of the campaigns. Now we have reached an “exclusive state” where the campaigns do not overlap.

As we keep adding more negative keywords to our phrase match campaign we need to ensure that we do not create situations where we reach an “exclusive but NOT exhaustive” state. This would lead us to accidentally exclude search queries that may be relevant for us (in this example “hooded baby towel”).

In practice this can easily happen, just think of products that come in different sizes and colors. One “filter” negative keyword can wipe out an entire search query word cloud.

Adding Negative Keywords at Campaign or Adgroup Level?

Negative keywords for Amazon can be applied to both campaign and ad group levels. The scope of their impact is determined by the level at which they are added.

When you add a negative keyword to an ad group, it just appears in that ad group; but, when you add a negative keyword to a campaign, it’s effective for every ad group in that campaign.

Adding a negative keyword to a campaign may have a significant influence on the campaign’s impression and its related ad groups. When adding negative keywords into the campaign level, you must be very selective.

The best practice is to add a negative exact match at the ad group level for the majority of the negative keywords. This guarantees that the scope of what is blacklisted is narrow and specific. There is some situation where we use negatives phrase match and/or on a campaign level, but those are cases where we know for sure we don’t want to be appearing for, a good example of this is gender-specific items.

How can SellerMetrics help?

As mentioned you can use the Amazon search term reports to optimize your campaigns by adding and negating search terms. But you can see that doing this manually is inefficient and prone to errors.

This is an especially tedious process when you are operating multiple marketplaces and advertising accounts.

We at SellerMetrics can either streamline this process or automate it entirely. Let’s see how it works!

1. Set the Rules

You can do this in the Automation Settings screen in SellerMetrics under “Search Term Rules” or “Negative Keyword Rules” columns.

2. Enter Rules Settings (Search Term Rules)

Here you will add settings to trigger search terms within a campaign. You will also tell the system the campaigns/ad group combo you want the search terms to be added to as a new keyword and enter what order threshold would trigger the search term.

Below are the settings to enter:

• Name of the Rule (used to call the rule being created)
• Destination Campaign (can select one to many)
• Destination Ad Group (from the campaign selected, can select one to many)
• Order/Max ACoS Threshold (tell system only grab search terms that satisfies the threshold)
• Target Type (Keywords or ASINs)
• Match Type and Bids

3. Add Search Terms as Keywords into Destination Campaigns

After adding the search term rules, the triggered search terms will show up on another screen for you to push the new keywords into the destination campaign with one click.

As mention earlier search term turns into a keyword in a campaign during this optimization. This is exactly how it is done!

4. Enter Rules Settings (Negative Rules)

Here you will add settings to trigger fo search terms to negate/blacklist within a campaign. You will also tell the system the campaigns/ad group combo you want the search terms to be added as a negative target and enter what click threshold with 0 order that would trigger. Below are the settings to enter:

• Name of the Rule (used to call the rule being created)
• Destination Campaign (can select one to many)
• Destination Ad Group (from the campaign selected, can select one to many)
• Click Threshold (tell system only grab search terms that satisfies the threshold)
• Automate (enabled the system to add search terms automatically)
• Target Type (Keywords or ASINs)
• Match Type (negative exact or negative phrase)

5. Add Search Terms as Negative Targets/Keywords into Destination Campaigns/Ad Group

After adding the negative keyword rules, the triggered search terms will show up on another screen for you to push the new keywords to the negative list into the destination campaign/ad group with one click.

FAQ: How to Find Negative Keywords for Amazon PPC Ad Campaigns