Posicionarnos SEM y Marketing en buscadores How to Do PPC Keyword Research in 2019

How to Do PPC Keyword Research in 2019


If you’re looking for the latest strategies in PPC keyword research, there’s something you should know: Google was never designed to be about keywords.

Google co-founder Sergey Brin said:

“My vision when we started Google… was that eventually you wouldn’t have to have a search query at all – the information would just come to you as you needed it.”

Google’s getting closer to that mission, and in 2018 it rebranded AdWords as Google Ads, dropping the “words” entirely. Many older keyword building methods are now outdated or defunct.

Today, there’s a new and easier way to handle PPC keywords, and it starts with a focus on users. To understand this new strategy and why it works better, it will help to know what’s different about the current search landscape.

Google’s Giving Less Weight to Keywords

Once upon a time, many algorithm updates ago, Google’s best chance of serving up relevant results was to match a user’s search terms with keywords on a page (or, in the case of paid ads, keywords in a list). A lot has changed.

Natural Language Processing Advancements

Last week, I did a Google search for a podcast episode. I couldn’t recall the episode number or name. But I remembered the gist of it, and Google knew what I meant.

Five years ago, Google wouldn’t have been able to deliver this result, but today Google’s natural language processing and AI have advanced far beyond simple word matching.

Google can now understand syntax, entities, sentiment, conversation, and context. Neural matching is being used for 30 percent of search queries. This makes it easier for people to get answers based on what they mean, not what they type or speak.

Personalization & Implicit Signals

Search for “dentist” and you’ll probably get a list of nearby dental offices. How does Google know to deliver those results (and not articles or definitions) without you even entering your city?

Google uses implicit queries (such as your location, search history, behavior, and demographics) to personalize the results you’ll see.

Combining your explicit queries (what you type or say) with implicit signals (who you are, where you are and what you’re doing) lets Google tailor the SERP with the most relevant answers, listings, and ads.

Taking the Keyword out of the Ad Serving Process

One of the driving forces behind Google Ads’ auction model is Ad Rank, which is based on factors like a keyword’s ad relevance.

But today, many ad formats that appear on Google Search don’t even have the option of bidding on keywords.

Campaigns for Dynamic Search Ads (DSA), AdWords Express / Smart Campaigns, Local Service Ads and Shopping Ads are completely keyword-less. They rely on your business type, products, or website to control when ads are served.

Google Changed How Match Types Work (or Don’t)

Google’s always been out to expand keyword coverage beyond the lists uploaded by advertisers. Here’s their official explanation for casting a wider net:

“There’s a good chance people are searching for your products or services with terms you haven’t discovered. Take deodorant, for example. Last year, we saw people search for deodorant in more than 150,000 unique ways. That’s a lot of different ways to say the same thing. But you shouldn’t have to manage an exhaustive list of keywords to reach these hygiene-conscious consumers.”

For more than a decade, Google let advertisers use broad match keywords to catch misspells and relevant variations of queries, while keeping tighter bid control with phrase and exact match types. This option has slowly been retired.

In the last few years, Google changed the function of exact match to include abbreviations, stemming, reordered words, function words, implied words, synonyms, and search intent:

Advertisers can no longer opt out of this coverage.

Now that exact match is behaving more like broad match, true exact coverage is lost. As the above Google quote implies, to show only for the exact query [deodorant], you’d have to add thousands of negative keywords.

With expanded keyword coverage implicit in every match type (like it or not), precise keyword control is no longer an option. Bloated keyword lists just add needless management complexity.

Google Lets Fewer Keywords Trigger Your Ads

Here’s a real screenshot from a keyword I created in June 2007:

The only impression came from my dad (it was part of a Father’s Day present), and he clicked the ad that led to a landing page with his gift.

Obviously, this keyword doesn’t work in 2019. Let’s look at why:

Low Search Volume = Inactive Keywords

Google marks keywords with few searches as “Low search volume,” which means they’re inactive and won’t trigger ads. According to Google:

“Keeping these keywords out of the ad auction helps Google Ads serve ads more efficiently and reduces the volume of keywords on our system.”

No Commercial Intent = No Ad

Around the time Google changed its Ad Rank calculation to weigh bids more heavily, it also updated Ad Rank to take into consideration the meaning of the query.

If this sounds complicated, think of it like this: Google’s trying to avoid delivering SERP experiences that feel spammy to users by restricting how often (and when) ads can show.

Even if advertisers are bidding on keywords, Google is less likely to serve ads if it believes the searcher is looking for information rather than a product or service.

Want to show your football t-shirt ads when someone searches “super bowl”? That’s not an option anymore. But you can show for “super bowl t-shirt” searches all day long:

Keywords that are unlikely to trigger a click (generating revenue for Google) aren’t worth its processing power. “Just in case” keywords are no longer eligible to serve ads and don’t benefit your account.

New & Improved Keyword List Building

All these changes add up, and Google Ads marketers now find themselves in unchartered waters. Today:

So how can we take advantage of the new keyword landscape?

There’s good news. While keyword research may be getting increasingly complex for SEO, it’s actually easier and faster now for PPC.

Here’s a user-first keyword research strategy that will improve your paid search campaigns and save you both time and money.

1. Define Your Goal for Paid Search

Don’t invest in PPC unless you know exactly how it will help to grow your business. I always like to remind advertisers:

“Paid search puts your best offer in front of your ideal audience when they’re most ready to take action.”

You’re not “buying traffic” or “buying keywords.”

You’re paying a premium to put your offer directly in front of people who are looking for it. What’s the offer your market is ready to say “yes” to (and take out their wallets for)?

2. Start with What’s Closest to the Money

Once you’ve defined the action your audience takes that drives leads and sales, you’ll identify who’s most ready to take that action.

We’re not making decisions about keywords yet, we’re making decisions about our audience and prospects.

In Breakthrough Advertising, Eugene Schwartz breaks out prospect awareness into five separate stages.

Applying stages of awareness to paid search, we can think of each stage as an electron shell – a nerdy analogy, but the best I could think of – with audiences increasing in size as we get further from the goal:

You can typically expect higher conversion rates and better performance from the “most aware” audience, so you’ll want to focus your efforts there first.

But you’ll eventually want to branch out – either because there’s not enough search volume for “most aware,” or you want to fill your customer pipeline.

3. Create Your Keyword Categories

Your next step is to start filling in keyword themes that support your audience’s stage of awareness.

This is not a one-size-fits-all exercise; you’ll bookend your categories based on the search volume, value, and specificity of your own products and services.

For example, “freelance copywriter” could be a product-aware term for a marketplace offering freelance copywriting services, but a problem-aware term for a copywriter specializing in long-form sales copy.

Most Aware

Your prospect already knows and wants your offer, and is ready to take action

Product Aware

Your prospect knows what you sell, but hasn’t decided who to buy it from

Solution Aware

Your prospect is looking for a result, but isn’t aware of specific brands or services

Pain / Problem Aware

Your prospect is trying to solve a problem, and either doesn’t know how or is considering many solution categories

4. Define the Right Keywords & Research Questions

Now that you’ve got a general idea of the keyword themes to target and you know who’s most likely to take action, what’s left to research?


But the best research doesn’t come from blindly auto-generating or “stealing” large keyword lists. It comes from listening to your prospects and digging deep into your own offer.

Mine Your Customers’ Language

Google can solve for a lot of synonyms, but you need to know how your market is thinking about your services. Especially if they are in the problem-aware stage, they may not have the vocabulary to describe your offer the way you do.

Zero in on the exact language your audience is using with review mining. You can use review sites (Amazon, Yelp, G2Crowd, etc.) and your own user data (Hotjar, surveys, chats, transcriptions) to add to your keyword list and find new terms for each stage of the funnel.

Get Clarity on the Offer

It can be tough to get a clear understanding of the products you’re marketing, especially when the industry is highly specialized.

Landing pages can be vague or full of jargon, making it hard to know what problems the product or service solves.

To get a better handle on what the offer is about, try these sources for neutral, non-branded language and keyword themes:

Once you’ve got clarity on your offer and stages of awareness, you can also use keyword research tools to identify and fill in gaps in your coverage.

Beware of the ‘Buyer Intent’ Party Trick

You may have heard that there are keyword modifiers that signal buyer intent (or lack thereof).

For instance, terms like “how to open a bank account” or “best running shoes” are informational/consideration terms, while “open a bank account” or “buy running shoes” are decision/transaction terms.

The truth is that most of the time, people don’t explicitly state their intent with boilerplate modifiers.

Remember our super bowl example above?

More importantly, the customer journey isn’t so binary.

If you search “luggage reviews,” read an article comparing luggage brands and find one you like, what’s your next step? You’d probably:

You wouldn’t start your search completely over with “buy luggage” and click an unrelated ad.

Putting It All Together

Let’s see how our PPC keyword research strategy works from start to finish. We’ll use a (made up) massage therapy school in Utah as an example:

Doing keyword research this way speeds up your process and helps you reach your audience when they’re most likely to take action.

A Few Final Tips

If you thought PPC keyword research meant you had to find 100,000 keywords to reach your audience: congratulations!

You can now let machine learning take over the grunt work of intent and keyword coverage, while you focus on strategy and searchers.

Coming Soon: Keyword Optimization

Once your keywords are live, the real fun begins. You can use the search term report to see the detailed performance of the actual search terms your audience is using (not just the keywords in your list).

The search term report will help you improve and curate your keyword list. But if you’re starting from scratch with keyword research, it’s unfair to suggest using a resource that doesn’t exist for you yet.

Use the steps from this article to build out your audience-focused PPC keyword list today. Then be sure to catch the companion article, How to Optimize Your Paid Search Keyword List, in May 2019.

This content was originally published here.

Leave a Reply

Tu dirección de correo electrónico no será publicada. Los campos obligatorios están marcados con *