Negative Keyword Lists Every eCommerce Brand Needs

In theory, PPC advertising is pretty easy. Pick some keywords, write up an ad, attach some images if the channel you're using allows them, and put money behind it. Watch how they work, tweak the keywords or ad copy, repeat until you're either out of money or you're successful.

Hidden below the surface is a lot more complexity than many people imagine. One, in particular, is the concept of negative keywords. What are negative keywords, how do they fit in PPC advertising, and how can you use them properly? 

What Are Negative Keywords, Specifically?

Negative keywords are keywords that you don't want your ads to appear on.

They are not necessarily keywords with negative sentiment. While you could add some of those to your negative keyword list, they aren't the main focus. 

It's easiest to demonstrate with a few examples. To start, say you're running PPC ads targeting your own brand name, so you're always sure to appear at the top of the list for anyone searching for your brand.

Now, remember what the goal of your PPC advertising is; it's to bring people in and convert them into customers. You pay money to put your link higher up in front of people's faces, with the expectation that some proportion of those people will click it, and some proportion of those people will make a purchase.

Now, say you're looking at your ad history, and you see the referring keywords people used where they saw and clicked your ad. You see things like:

• [Brand Name] Careers
• [Brand Name] Free Plan
• [Brand Name] Cancellation
• [Brand Name] Scam
• DIY [Brand Name]

All of these are, quite possibly, people who click through to your site. That click costs you money. Yet, none of these people are going to be making a purchase. Some have other intents (careers), some have no intention of paying money (free plan, DIY), some are already customers (cancellation), and some are still in the research phase (scam).

You paid to get those clicks, but there was a 0% chance that those clicks would turn into conversions.

Negative keywords are the answer to this problem. By including keywords that are sure signs of non-commercial intent, you can prevent your ads from showing for those queries.

What's important is that this does not prevent those people from finding the information they were looking for. It doesn't prevent your site from showing up in the search results, nor does it prevent them from clicking through to you. What it does is prevent your ads from wasting money.

Negative keyword lists can get very long, and that can end up with a lot of potential problems. So, how do you use negative keywords properly?

How to Use Negative Keyword Lists the Right Way

Using negative keywords is simple on the surface. All you need to do is build the word list, add it to the negative keywords list field in your paid ads (almost all search-based PPC ad platforms have this option), and profit.

The problem lies in the complexity of the lists more than anything. So, here are my tips and best practices. Don't worry; I'll get to specific lists later in the post.

Understand the Way Your Platform Uses Negative Keyword Lists

The first thing to do is understand exactly how the platform you're using actually uses negative keyword lists. 

One big detail is that negative keywords generally only exist on ad platforms that rely on user searches. Google is the most obvious, but Apple uses them, as does Microsoft, and more. 

This is because they rely on keyword matching from user queries. That's easy when a user types in something they want to learn about, but it's a lot harder for other kinds of ads.

• Display ads on websites. You have a much harder time using negative keywords in a system that displays ads on webpages instead of search results. There are controls here (so you can, for example, not have your ads display on a competitor page), but they aren't nearly as granular or robust. There are also publisher-side controls to contend with as well.
• Display ads on social media. Similar story here; since a user isn't making a query, ads aren't filtered by keyword. Facebook/Instagram/Meta is the big example here; they don't have negative keyword options. Instead, it's all about narrowing the demographic and interest targeting.
• In-App ads. Since these are display ads generally based on broad interest categories, and there's usually no user interaction beyond tapping an X, there's no real opportunity for negative keywords here.
• LLM Search ads. Some of the LLMs are starting to roll out advertising in their responses. ChatGPT was one of the first, and I've started to see sponsored recommendations in Google's AI Overviews, too. Since LLMs don't work on traditional keyword matching, these don't really use negative keywords in the same way. Google's might (I haven't checked from their end yet), but ChatGPT doesn't.

I highly recommend looking for documentation on how the platform you're using actually uses negative keyword lists, as well. Think about whether or not they allow multiple lists or just one, if there's a maximum number of keywords per list, what the match types are, and more. Here are a few of the help documents to show you what I mean.

• Apple App Store Ads
• Google Ads
• Microsoft (Bing) Ads
• Amazon Ads

When you read through these, you can see that there are subtle differences that can matter. Make sure you know the specifics for the ads you're running.

Maintain Multiple Lists

This tip is a two-in-one.

First: maintain multiple lists with different purposes. When you just have one massive list of negative keywords, a few things happen.

• It becomes huge and unwieldy to manage.
• You can accidentally add the same keyword more than once and not notice.
• You can run into limits on the size of a list more easily.
• You can end up with a lot of negative keywords that aren't relevant to a particular ad campaign.

I like to have a handful of lists for specific purposes. For example, you might have a list for DIY/Educational keywords you can use for products of an instructive nature, a list of career-oriented keywords you can use for non-career ads, and a more general list of low-intent modifiers like "cheap" or "used" or "free". I also like to have a list of competitor-relevant keywords that would indicate an adversarial intent, as well as a list of task-oriented keywords like "support" or "contact" that the organic search will be better for serving up, and that you don't want to pay for.

Basically, treat your lists like modules in a larger superlist. You can add and remove modules without having to retool an entire list for every campaign.

This also allows you to leave out chunks of lists for specific ad campaigns. For example, if you're running a campaign specifically targeting your competitor's brand keywords, you don't want to have those keywords in the negative list.

The second part of this tip is this: treat the lists as living documents. You aren't building one big list and calling it a day. You'll need to modify these lists on an ongoing basis. Add new keywords, shuffle keywords from one list to another, and so on. 

Google, at least, makes this easy. In your Google Ads reports, you can run keyword reports for your ad keywords and see which search terms were used to display the ads. You can easily identify the ones that aren't relevant and, right there in the dashboard, add them as negative keywords. This is just the internal list with Google, but you can also export that list and save it for later.

Don't Just Blindly Download and Use Online Lists

As you'll see later or if you've searched for this topic before, you can easily find prepopulated lists of negative keywords that can be thousands long. It's tempting to just download and start using these lists. But if you don't know what's in them, you can run into all kinds of problems.

One of the biggest, for example, is "free" keywords. These can be important for some brands to flag as negative, because they indicate low intent to pay, so they aren't worth paying for. But what if you offer a free trial or free tier of your service? Cutting out anyone who would be interested in that (and thus who could convert down the road as an upgrade) isn't smart.

More than once, when troubleshooting why client ads aren't showing the way they should, I've found negative keywords to be a primary culprit.

Now, I'm not saying not to use lists you find online. I'm just saying to review them before you upload and use them.

Which Lists Are Essential for eCommerce Businesses?

Now, let's talk about more specific lists. I'm not going to publish whole lists here, but rather give categories and examples to get you started. You can then find larger examples online, use your favorite LLM to spin one out, or just start small and let them grow over time.

Personally, I usually go for the latter option. When you start small, you may waste a little money on ads, but you have far fewer issues with conflicts. 

List 1: Careers and Education

Careers and education is an important list to build for two reasons. On one hand, it's an audience that definitely wants to find you and the relevant pages on your site, but you don't want to be paying them to click through for it. Unless you do; sometimes, you might want to advertise a hiring push, and then you can convert that negative list into a positive list.

Keywords on this list include things like jobs, careers, hiring, salary, resume, interview, internship, employment, recruiter, and names of job sites like Indeed, Monster, Glassdoor, LinkedIn, and so on.

The education side can be the same list or a list of its own. It's for people who want things like tutorials, courses, certifications, and programs related to your product or industry. This may or may not even be relevant to you.

List 2: Cheap and Free

Cheap and free lists are all about the people who want your product, but don't want to pay for it. These can range from looking for coupon codes or sale dates to hoping to find cracked versions of a program, sideloaded APKs, shared account information, or illegal copies.

Bear in mind that some of the keywords you might add to this list can still be useful. If you offer a free trial, blocking "free" means your ads don't show up to people looking for it, which can be a damper on how many sign up. That may be a good thing if you have a very low rate of upsales, or it can be a bad thing if you use that as a gateway more often than not.

Keywords for this list can include words like coupon, deal, sale, discount, crack, keygen, free, and even filetype searches, particularly for digital products.

List 3: Out-of-Bounds Locality

This may or may not be an important list for you. Geographic targeting and local search are hugely important for certain businesses, and completely irrelevant to others. Moreover, ad managers also have robust geographic targeting built in as well. You may not need to exclude geographic locations with your lists.

I find this most relevant for local businesses that have geographic areas a little harder to define than simple city/county/state names. For example, a client with a service area that includes half of New Jersey could exclude the counties and major cities in the other half. 

List 4: B2B (or B2C)

A B2B business probably doesn't want retail customers, and a B2C business isn't necessarily equipped to handle business contracts. By including keywords indicative of those kinds of operations, you can minimize wasted ad spend.

Keywords on a list like this can include terms such as wholesale, bulk, dropship, white label, reseller, affiliate, and overstock.

List 5: Resources and Tasks

This category is something you still want people to find on your site, but you don't want to be paying for them to find. Either search on its own will find it, or your own navigation or site search will. Either way, you don't need paid ads pointing to landing pages for it.

This is where you use keywords like support, refund, complaint, feedback, contact, knowledge base, tutorial, and other tasks. Things users need to do on your site, but aren't commercial.

List 6: Profanity, Adult, and Spam

I don't know how many people actually type profanity into a search and expect anything out of it. I do know that it's getting more common, both as a way around certain AI Overviews (though it doesn't work as much anymore) and as a way to express frustration with LLM search.

Adult and spam keywords are also relevant to include, though again, presumably not many people are searching your ecommerce site for XXX content, adult dating content, or whatever else. 

I use this category as a catch-all for keywords that are 100% never going to be relevant, and only cropped up because someone was trying to be malicious with it. If they show up in keyword or traffic reports, it's an auto-include, but otherwise I don't go out of my way to populate it too deeply.

List 7: Product Variations

A critical list for ecommerce, you want to identify near-but-not keywords for products you don't sell. 

For example, if you sell casual and dress shoes, you don't want people seeing your ads when they search for work boots, ski boots, or other footwear you don't sell.

A common example I see is men's and women's as modifiers, but this isn't actually relevant in practice most of the time. How many shoe stores are so heavily gendered? I don't know of any. Other apparel, maybe, but even that tends to be rare. 

List 8: Off-Site Content

This is potentially one of the more useful lists. People will often search for your brand name, product, or service, but append another site's name to it. [Brand Name] YouTube or [Product] Udemy, for example.

These kinds of queries indicate someone who wants to learn about your product or brand, look up reviews or exposes or tutorials, and who doesn't expect the content to be on your site. The other site's name is an expression of intent to visit that other site, not your site, so paying to show your ads there isn't going to do you much good.

There are plenty of other lists you can build, too; this is just an example of how I've broken down and categorized my lists.

Making Negative Keywords Easy

There are a couple of things you can do to make building and managing negative keyword lists easier.

The first is to look up, copy, and analyze existing lists. Sites like Level, AdWhiz, AdLabz, and even user-generated lists posted to Reddit can all be good resources here.

You can also use LLMs to spin up big lists of negative keywords. Since they're trained on existing lists, they can produce reasonably thorough lists. However, make sure to vet every word on an LLM-generated list. LLMs are pattern matchers, not expert sources, so they can easily include keywords you would find essential. 

Finally, I've built a couple of free content marketing tools, mostly for my own use but which I made public for anyone to get value out of.

The first is my negative keyword list builder. This lets you upload, manage, import, and categorize your negative keywords. The second is my negative keyword conflict checker, which is a stand-alone version of one feature of the first tool. You plug in your negative list and your bidding list, and it checks if there are conflicts.

Both of these are completely free and can take a lot of hassle out of managing your lists, especially once you have multiple spreadsheets of hundreds or thousands of keywords. 

Any questions? I'm happy to discuss more; just let me know in the comments.