The importance of targeting branded searches

Search experts understand the importance of targeting non-branded search terms: Optimizing for high-volume, non-branded terms can drive a significant amount of traffic to your brand’s site.

While targeting non-branded search terms plays an essential role in your overall search strategy, many brands still underestimate and neglect the power of branded search terms. By relying on the strength of a brand and integrating branded search tactics with current non-branded search strategies, your business can discover more qualified leads — and, as a result, increase conversions.

The role of branded search in consumer behavior

For many established companies, their branded terms make up a majority of their keyword profile. If people are searching for your brand or products by name, they’re likely deeper within the sales funnel. In fact, Google has found that branded keywords have over two times higher conversion rates when compared to non-branded terms. So why would brands shy away from increasing or stepping up their branded search efforts?

Let’s flip the script and put you in the customer’s shoes. Say you’re searching for a fitness tracker your brother would love this holiday season. When you begin your gift hunt, are you more likely to search for “best fitness tracker for men,” or for “best Fitbit for men”?

Data from Google AdWords Keyword Planner

Due to the Fitbit’s brand awareness efforts, the product is iconic enough that consumers search for it more often than non-branded terms. Search engines like Google, Bing and Amazon recognize the strength of the brand — SERP layouts and competitive pricing reflect this.

Brands working to improve their conversions need to work the entire marketing funnel. For brands or products well-known enough for branded search terms to be relevant to audiences, it’s important to know how audiences discover your products so you can target these branded search terms. Otherwise, you’re leaving money on the table for competitors and review sites to take for themselves.

Integrating branded search tactics into marketing strategy

With a strong brand, and thus stronger branded search terms, bread-and-butter search tactics will have some incredible advantages. These advantages span both paid search and organic search tactics, affecting every aspect of search from the page rankings, search boxes, knowledge panel, and even map results. With branded search, search engines will recognize your main site, if optimized for best practices, as the most relevant site for searching by potential customers.

Organic search

Your home page and (if applicable) product category pages should rank the best for high-traffic branded search terms. Your title tags and meta descriptions should clearly display these branded search terms and relevant context that encourages searchers to make the decision to click. Once they do, the site should match the promise the SERP listing made with this copy.

Your goal should always be to dominate the first page and to obtain the highest positions with optimal branded search efforts. Brands should not only focus on their branded terms at the user’s research and consideration phases of the funnel, but also the post-purchase phase.

In the research phase, searchers will find strong, relevant brands first and foremost. Consider that they will also be looking for reviews, pricing and where to buy the product. This information should be available to users prior to their converting.

But the job isn’t done after converting. Post-purchase, many users will search for more information about the product using branded terms — installation instructions, how-to guides, proper cleaning and maintenance techniques, general product help and more — and these searches should lead to your website.

All of the brand-related terms throughout the sales funnel have heightened search term relevance to affect consideration, conversion and continued use.

Paid search

If you plan to own as much real estate on the SERPs as possible, paid search is an essential tactic to earn qualified leads. Even if you have obtained the top ranking in organic search results, research suggests that having an ad can produce incremental clicks. With little competition, it’s pretty easy (and cheap) to own these paid search spaces.

With branded search ads, you should be making use of ad extensions. These will provide more information to searchers, which can make your ad stand out and entice users to click. Certain extensions — such as sitelink, location or price extensions — can also increase your listing’s SERP real estate, particularly on mobile.

Keep in mind that Google factors ad extensions into its Ad Rank calculation, so proper use of extensions can give you an edge over competitors who may be running conquest campaigns on your brand name.

Other branded search considerations

Beyond the basics of SERPs with organic and paid search listings, you should be taking advantage of additional branded search real estate options that should be taken advantage of, including:

Organic sitelinks, the links that are displayed under the top organic search result. They’re important since they occupy a lot of SERP real estate and can function as an outline of your site, helping users to navigate to your other top pages. Google determines whether it will provide sitelinks or not, so you don’t have direct control over this — but you can help Google out by submitting an XML sitemap and having your site set up with a logical hierarchy.

Apple shows six organic sitelinks for a branded search. Note the site search box, too.

You want to make sure map results are showing up for you if your brand or business has physical locations. To do this, you need to ensure that you’ve set up Google My Business listings with the correct NAP (name, address and phone) information.If your site has an internal search function, you then have a solid chance of a search box showing up on Google. If it doesn’t display within the SERPs, you can utilize structured data markup per Google’s guidelines.The Knowledge Graph helps users discover business information quickly and easily. Google will pull this information automatically from trustworthy sites like Wikipedia or WebMD. With the right mix of search tactics, you can obtain a Knowledge Graph result for your brand. Make sure that you have all social channels, a solid description, reviews and accurate information correctly displaying in the eye-catching Knowledge Graph.

Final thoughts

Branded searches are imperative and shouldn’t be overlooked. Many assume that search queries involving your brand will naturally lead to your website, but that’s simply not the case. Without optimizing your paid and organic search efforts to capture branded searches throughout the entire purchase cycle, you’ll be missing out on tons of potential new traffic and conversions. Owning as much real estate as possible for your brand is crucial, especially during high-traffic seasonality.

Oftentimes, branded search terms can be the last channel touch point for consumers who are about ready to convert on one of your products or services. By incorporating branded search into your overall digital marketing strategy, you can quickly accelerate your brand, helping it stand out on the SERPs and provide a better experience to audiences.

Opinions expressed in this article are those of the guest author and not necessarily Search Engine Land. Staff authors are listed here.

How to steal the competition's best keywords: A 3-step guide

Whether you’re new to paid search or an old pro, one of your biggest challenges is figuring out the right keywords. Pick the right keywords and your business will thrive. Pick the wrong keywords, and you might as well be flushing your ad spend down the toilet.

Now, with enough time, effort and money, most advertisers can usually put together a decent keyword list. The only problem is, this takes time, effort and money — something most businesses are a little short on.

Wouldn’t it be nice if there were a way to make the process more efficient?

The good news is, while identifying the right keywords won’t ever be easy, there is a quicker way to find the right keywords for your business. All you have to do is steal them from the competition.

Learning from the competition

If you’ve got competitors — especially successful, well-established competitors — there’s a strong chance that they’ve put a lot of blood, sweat and tears into finding the right keywords for your market. Odds are, they’ve got a good feel for what works and what doesn’t.

Wouldn’t it be nice if you could get that information?

Fortunately, there is a way. With the right approach, you can quickly identify the competition’s best keywords and beat them at their own game. All takes is the right analytics setup, a competitive analysis tool (iSpionage and SpyFu are good options) and a testing budget.

1. Analytics

The key to any paid search strategy is great analytics. After all, how do you know which keywords are producing the most profitable conversions and sales if you aren’t, you know, tracking conversions and sales?

So, before we dive into identifying your competition’s best keywords, let’s take a moment to talk about the analytics information you’ll need to make this strategy work.

Both Bing and AdWords make it pretty easy to set up conversion tracking. The easiest types of conversions to track are form submissions and online transactions, but odds are you’re not just interested in form submissions — you want people to chat in, call you, make an offline purchase and so on.

Each of these actions is a conversion and an important indicator of how a keyword is performing. So, if you want to know which keywords produce the best results, you need data on every single one of those conversion actions.

Getting all of this set up can be a bit of a headache (especially if a lot of your conversions happen offline), but if you aren’t tracking the results of your paid search campaigns, you’ll never be able to improve your performance.

At first glance, this might seem kind of obvious, but after auditing over 3,000 AdWords accounts at Disruptive (disclaimer: my company), we’ve found that less than 30 percent of advertisers are actually tracking all of their important conversion actions.

In our study, AdWords advertisers who were only tracking some of their conversion actions were missing out on more than a third of their conversions. Just imagine what it’s like trying to optimize your campaigns with a margin of error of 33 percent!

With all that in mind, it should be obvious why analytics is the key to this tactic. Stealing the competition’s keywords is great, but if you really want to know which keywords produce results, you need analytics.

2. Competitive research

Advertisers love to talk about competitive research, but it’s often hard to know what is working for your competitors and what isn’t. Does their success come from their ad copy? The testimonials on their landing page? Or just the keywords they’re bidding on?

In many cases, you’ll probably never know. You can take your best guess, but you really have no idea what is working for your competitor and what just looks nice to you. More importantly, even if a certain catch-phrase or testimonial is the key to their success, you can’t assume that the same elements will work well for you.

So, while competitive research sounds nice, it’s often hard to use in a practical way.

When it comes to keywords, however, competitive research is incredibly practical. With the right competitive analysis tool, you can know exactly which keywords the competition is bidding on. In most cases, all you have to do is hop on a site like SpyFu or iSpionage, enter the competitor’s domain, and you’ll get a list of all the keywords your competitors are using.

From there, all you have to do is identify which keywords actually produce value for your business.

Unfortunately, most advertisers bid on way too many keywords. In fact, while auditing all those AdWords accounts I mentioned earlier, we found that only 6 percent of the keywords in the average AdWords account are producing conversions.

So, if you just assume that every keyword the competition is bidding on is profitable and use it in your own account, you’re going to end up wasting just as much money on paid search as your competitors are. To really make the most of this strategy, you can’t just copy and paste a competitor’s keyword list. You have to beat them at their own game.

3. Stealing the competition’s best keywords

If you want to steal the competition’s best keywords, you need to know which keywords are worth bidding on and which ones you (and probably your competitors) should ignore.

That’s going to take a bit of testing.

I typically recommend that you invest about 20 percent of your paid search budget into testing new ads, keywords or other changes to your account. Obviously, the newer you are to paid search, the more of your budget you’ll want to test with, but this is a good rule of thumb for most established accounts.

Once you’ve identified the competition’s keywords, build ads around them and use your testing budget to try them out for around three months. Assuming that you have a good analytics setup, this should give you enough data to determine which keywords are profitable.

Next, open your paid search account and take a look at the three months of data you just gathered. Go to the Keywords tab and filter your keywords for Conversions < 1.

This report will show you all the keywords you are bidding on that haven’t produced conversions in the last three months. By clicking on the “Cost” column, you can further sort these keywords by how much money you’ve spent on these keywords.

Now, if this weren’t a test, all the money you spent on these keywords would have been completely wasted. However, the point of your testing budget is to identify useful keywords, so you’re not wasting money — you’re just paying for education.

You’ll have to use your individual judgment to decide which of these keywords should be cut and which deserve a little more time or refinement. But, this filter should help you identify which of your competitor’s keywords are producing unsatisfactory results.

In contrast, you can reverse this filter (“Conversions > 1”) to see which keywords are producing results. These keywords are often your competition’s best performers, too. However, since they probably aren’t tracking conversions effectively, they probably don’t know how valuable those keywords are.

But you do.

Once you’ve identified a great new keyword, all it takes is a little more time and effort to dial in your bids, ad copy and landing pages and edge out the competition. Guess what? You just beat the competition at their own game!


When you get right down to it, it’s not that hard to steal the competition’s best keywords. All you need is a good analytics setup, a competitive analysis tool and a decent testing budget.

By combining these three elements, you can quickly identify which of a competitor’s keywords are valuable to your business and which ones aren’t. Then, you can focus your ad spend on their best keywords and beat the competition at their own game!

Opinions expressed in this article are those of the guest author and not necessarily Search Engine Land. Staff authors are listed here.

The PPC challenge of selling manufacturing capability vs. stock products

While running a successful advertising campaign can be complex at the best of times, it presents a special challenge to small industrial manufacturers.

It’s one thing to advertise a widely understood product, such as a running shoe or blender, to a large consumer market. It’s another thing entirely to sell a manufacturing capability that’s only really understood by those in the industry — especially when products look and function differently based on the specifications of the buyer.

This challenge is particularly pronounced in the world of search engine marketing, where industrial manufacturers must adapt to an ad platform (e.g., AdWords) that seems more geared to retailers.

My company, Huff Industrial Marketing, faced this challenge with a manufacturing client. I’m pleased to report that through carefully and patiently running a variety of experiments, based on different approaches and ideas, we’re now seeing good results.

In this column, I’ll share how we did it.

A manufacturer of specialized machine parts

The client is in the business of making specialized machine parts used in a variety of manufacturing processes. Because each part is custom in size, type, material and so forth, what this client is really offering is the capability of making these specialized products. In other words, they don’t make “stock” products sold off the shelf.

This immediately created a conundrum with the PPC advertising campaign — namely, how in the heck do you advertise a manufacturing capability to buyers in dozens of industries?

Eventually, we came up with a solution that encompassed three critical PPC elements: keywords, messaging and landing pages.

Element #1: Keywords

Almost immediately, we knew that keywords were going to be a problem. One challenge is that terminology used in manufacturing is often specialized and used as “jargon” inside a company or industry niche — but may not be fully understood or used by buyers across the myriad industries or markets that exist.

Consequently, searches on these terms often return irrelevant results.

For example, when I conducted a search on a specialized type of manufacturing wheel, it returned results for car wheel rims and bike wheels — which wasn’t at all what I wanted (Figure 1).

Figure 1: Google keyword search result

Unfortunately, this didn’t seem to be something that Google’s machine learning could correct.

So, what to do? When we went too broad with our keywords, the ads ended up displaying with irrelevant search results. (Of course, we worked to mitigate this with negative keywords.) If we went too narrow, then search volumes were too low.

Eventually, what saved us was monitoring search queries for both organic search and PPC. Based on the data, we realized we had to stop thinking in terms of “capabilities” and start thinking in terms of “end products” — even when those end products weren’t standard.

Then, we marketed each end “product” individually through separate campaigns and ad groups.

As an aside, the client, too, was surprised at the terms buyers were using to source specific types of products. It’s been a real eye-opener for everyone on the team and has led to the client looking at new ways of marketing custom items they’ve been manufacturing for years.

Element #2: Ad messaging

We also took a closer look at our ad messaging — in particular, the calls to action.

If you’ve looked at PPC ads generated by industrial manufacturers, you’ll know it’s not unusual for these ads to include calls to action like “Free RFQ” or “Get a free quote!” in the ad body or sitelinks.

Figure 2: RFQ in ads

Initially, we followed the lead of other manufacturers and included similar calls to action in the ads. We realized very quickly they weren’t converting. When we thought about it more deeply, we realized that this kind of messaging didn’t make sense.

Why? Because at this point in the sales process, the client’s potential customers weren’t ready to embark on a lengthy quote process. They simply wanted an answer to the question, “Can you make this thing?”

It was a classic case of: “Don’t ask your prospects to marry you on the first date!”

In other words, prospects needed to talk to someone. So, we revised the ad copy — and landing pages — accordingly.

Element #3: Landing pages

Once we identified the issue with the call to action, we knew we needed to revise the landing pages. Instead of a call to action to fill out the RFQ form, we replaced it with a much shorter form that allowed potential customers to ask their “can you make this?” question (Figure 3):

Figure 3: Landing page form

At the same time, we reworked the landing page to remove any unnecessary navigation and competing calls to action.

This task was made significantly less onerous with the help of our web developer, who was able to reconfigure the client’s code and add simple “click to remove” options. This allowed us to remove the main nav, footers and so on by simply checking a box.

With this capability, we didn’t have to rely on a third-party landing page service, which saved the client money and also kept the landing pages on the client’s domain.

To top things off, the development team added the script for Google’s phone call conversion tracking, where Google dynamically changes the phone number on the landing page, allowing for tracking of desktop calls.

Result: Relevant traffic and good conversions

With these changes in place, we were quite pleased to see four “in-the-ballpark” form inquiries and four phone calls in the first five business days. Inquiries have continued to stream in at a good pace.

In conclusion…

What can industrial manufacturers take away from this experience? Three things:

    Focus on your business and sales process versus what other manufacturers are doing. It’s very easy to fall into the trap of “everyone else is doing it so it must be working.” Nope! Study your data and then experiment with different approaches until you see the results you want.Rethink what you’re selling. For many small industrial manufacturers, what is being sold is a capability — often to create highly specialized parts that go into bigger systems. Monitor organic and PPC search queries to see what terms potential buyers are using to find you — and let those queries be your keyword guide.Don’t give up! It’s very easy to become frustrated with AdWords (believe me!) when you’re spending money and seeing little, if any, return. Continually read the AdWords help files to understand how things work, find a good developer who can help you with tricky things like coding and tagging, and experiment. With patience, you will find a combination that works for you.

Opinions expressed in this article are those of the guest author and not necessarily Search Engine Land. Staff authors are listed here.

Keyword tiering: A systematic way to juggle your AdWords priorities

Ever feel like AdWords is one giant juggling act? You cut your cost per click, only to watch your conversion rate drop. You increase lead volume, but those leads don’t turn into sales. You finally boost sales, but now your cost per click is unprofitably high…

When does the madness stop?

With all of these apparently conflicting priorities, it can be easy to drop the ball a few times. Or, worse still, you might feel like AdWords isn’t worth the effort, throw your hands in the air and quit.

However, there is a way to juggle all of your marketing goals — a way to balance clicks, conversions, sales and return on investment (ROI).

The secret to this juggling act is… keyword tiering.

Keyword tiering

The idea behind “keyword tiering” (coining a new term here, I guess) is fairly simple: you categorize your AdWords keywords into different tiers based on the results they generate. Then, as your business needs change, all you have to do is turn tiers on and off to meet your goals.

So, if leads are more important than sales today, all you have to do is turn on a few more tiers. If profitability is more important than sales volume tomorrow, turn off a few tiers.

Once you have effective keyword tiers in place, you can instantly adapt to the needs of your company. It’s a straightforward way to match your AdWords advertising to your business goals. And the good news is, it’s fairly easy to implement:

Step 1: Make sure you’re tracking the right metrics

As great as keyword tiering is, it does come with one caveat. In order to set up meaningful keyword tiers, you need to know which keywords are producing which results.

If you’re optimizing for leads, you need to be tracking leads. If you’re optimizing for sales, you need to be tracking sales. If you’re optimizing for ROI, you need to be tracking revenue. Depending on which metrics you’re tracking, that may mean you need to implement conversion tracking on your site or ensure that your CRM is tracking sales at the keyword level.

Unfortunately, only 29 percent of AdWords advertisers are even tracking conversions effectively enough to use keyword tiering. Far fewer are tracking sales and revenue at the keyword level.

If that describes your business, now is the time to fix things. A full description of how to track AdWords performance through to sales is beyond the scope of this article, but this article should help you get started.

Step 2: Define your keyword tiers

Once you are tracking all of your relevant metrics and have keyword data for a couple of sales cycles, you can define your keyword tiers. How you define your tiers will depend on your specific business and marketing goals, but it’s often easiest to start with four basic tiers.

For example, if you are using AdWords to drive leads for your business, you might set up your tiers as follows:

Tier 1: Highly profitable keywords

Tier 1 keywords are your all-stars. They have a proven track record of producing profitable sales — in fact, there are very few reasons why you should ever turn these keywords off.

These keywords should always get budget priority and should never have their impressions limited by budget. If a Tier 1 keyword isn’t profitable above a certain average ad rank, it’s okay to lose impressions to ad rank, but never let these keywords lose impressions to budget limitations.

Remember, Tier 1 keywords are always profitable, so if you aren’t bidding on these keywords, you are losing money.

Tier 2: Break-even to slightly profitable keywords

Your Tier 2 keywords aren’t quite as reliable as your Tier 1 keywords. They might deliver hit-and-miss leads or mildly profitable sales, but you can’t count on these keywords to always deliver.

There are a few different categories of keywords that can end up in Tier 2:

Promising keywords. These keywords look like they’ll be winners, but you don’t have enough data yet to be certain. For example, you may have a few profitable sales from these keywords, but you’ve also only had a few conversions, so it’s too early to tell.“Swingin’ for the fences” keywords. Some keywords aren’t particularly profitable most of the time, but every so often they deliver that insanely good lead or sale that makes it all worthwhile.Reliably average. You’ve got plenty of data on these keywords, but you can’t seem to make them produce at a Tier 1 level. However, you can count on them to be somewhat profitable.

The good news is, you can often turn a Tier 2 keyword into a Tier 1 keyword with a little time and effort. But until that happens, keep your Tier 2 keywords in Tier 2 — no matter how much potential you see in them.

Tier 3: Keywords with potential

Tier 3 keywords are usually more bark than bite. These are the keywords that produce a lot of interesting results (like good lead volume or putting a product in a shopping cart), but that’s as far as things typically go.

As a result, these keywords have potential, but since you haven’t really figured out how to capitalize on that potential, they aren’t the first place you should be spending your advertising budget.

However, if your business is trying to grow and you’ve already maxed impressions for Tier 1 and Tier 2, Tier 3 can usually get you in front of a lot of potential customers. All you have to do is figure out how to turn that potential into actual revenue.

Tier 4: Long-shot keywords

Sometimes, you simply need more clicks and conversions. Maybe you’ve got a sales team that is starting to lose momentum due to low lead volume. Maybe your business needs more sales to bring on potential investors — even if those sales aren’t particularly profitable.

Business is complex, and sometimes volume is more important than profitability.

Tier 4 keywords are a perfect fit for these sorts of situations. In general, Tier 4 keywords seem like a borderline lost cause (you’ve had lots of leads but no sales from these keywords). That doesn’t mean they can’t produce good results, but the odds aren’t in your favor.

In general, I prefer to leave Tier 4 keywords off most of the time, but it’s good to have them on hand for the occasional time when you really, really need to boost clicks and/or conversions.

Testing tier

Okay, so this isn’t technically a tier, but I wanted to point out that new keywords you are actively testing should not be grouped into your regular tiers. The whole point of your tests is to determine whether or not a new keyword is viable, so you can’t exactly assign new keywords into a tier until you’ve finished your test.

Step 3: Create your tiers

Of course, simply tiering your keywords in a spreadsheet somewhere doesn’t do your business a whole lot of good. To really use your tiers, you have to be able to sort your keywords in AdWords.

To do that, you need to label your keywords.

Hopefully, you’ve used labels before, but just in case this is new to you, let me show you how to label your keywords by tier. First, open your AdWords account and click on “Keywords.” Check all the boxes next to your Tier 1 keywords:

Then, click on “Labels” and check the box next to the tier you want to assign your keyword to.

If you haven’t created your tier labels yet, click “Create new” and do it now. Guess what? You just assigned all of your checked keywords to Tier 1!

Now all you have to do is repeat the same process for all of your other tiers (you can also label keywords using the AdWords editor if you want to speed things up a bit).

As a quick side note, to see which keyword is assigned to which tier, you’ll need to make sure you’ve added the “Labels” column to your Keywords report. To do that, click “Column,” click on the double arrow by “Labels,” and hit “Apply.”

Congratulations! You just set up keyword tiering! Not too bad, eh?

Step 4. Use your tiers

Now that you’ve set up your tiers, it’s time to take your AdWords juggling skills from this…

To this…

First, you’ll need to create and save a filter for each keyword tier. If you’re still on the Keywords tab, click “Filter” and “Create filter.”

Set the drop-down menus in the filter to “Labels” and “contains any.” Then, click on the last drop-down menu and choose Tier 1 (or whatever you named your tier).

Give your filter a name, make sure “Save filter” is checked, and click “Apply.”

You can now pull up all of your Tier 1 keywords simply by running your custom filter. Use this same process to create filters for your other tiers, and you should be good to go.

Now that you have your tiers set up, the next time your boss says, “We need to cut our AdWords budget by 50 percent this month,” all you have to do is run your Tier 3 and Tier 4 filters, check the box to select all, and click “Edit” and “Pause.”

In no time flat, you’ve cut ad spend to all of your poorly performing keywords. You might still have to make some adjustments to get to that 50 percent budget cut, but you know that you’re spending that limited budget where it matters most.

This tactic might seem fairly straightforward, but the ability to quickly maximize AdWords performance in any given situation can have a huge impact on your business.

Just to show you how this works in real life, here’s what happened when we set up keyword tiering for a lead gen client last year:

In just three months, their cost per conversion dropped by 62 percent. To make things even better, this benefit filtered through to their cost per sale and tripled the profitability of their AdWords account!


If you set it up right, keyword tiering can allow you to juggle priorities in your AdWords account with ease. As a result, your AdWords campaigns will perform better, and your life will be easier.

Talk about a double win!

So, if you’re sick of dropping the ball on AdWords, try setting up keyword tiering. You might not end up in Cirque du Soleil, but life will certainly be easier!

Opinions expressed in this article are those of the guest author and not necessarily Search Engine Land. Staff authors are listed here.

Coaxing smarter paid search bidding decisions out of sparse conversion data

Paid search is an industry that’s grounded in data and statistics, but one that requires practitioners who can exercise a healthy dose of common sense and intuition in building and managing their programs. Trouble can arise, though, when our intuition runs counter to the stats and we don’t have the systems or safeguards in place to prevent a statistically unwise decision.

Should you pause or bid down that keyword?

Consider a keyword that has received 100 clicks but hasn’t produced any orders. Should the paid search manager pause or delete this keyword for not converting? It may seem like that should be plenty of volume to produce a single conversion, but the answer obviously depends on how well we expect the keyword to convert in the first place, and also on how aggressive we want to be in giving our keywords a chance to succeed.

If we assume that each click on a paid search ad is independent from the others, we can model the probability of a given number of conversions (successes) across a set number of clicks (trials) using the binomial distribution. This is pretty easy to do in Excel, and Wolfram Alpha is handy for running some quick calculations.

In the case above, if our expected conversion rate is 1 percent, and that is indeed the “true” conversion rate of the keyword, we would expect it to produce zero conversions about 37 percent of the time over 100 clicks. If our true conversion rate is 2 percent, we should still expect that keyword to produce no conversions about 13 percent of the time over 100 clicks.

It isn’t until we get to a true conversion rate of just over 4.5 percent that the probability of seeing zero orders from 100 clicks drops to less than 1 percent. These figures may not be mind-blowingly shocking, but they’re also not the types of numbers that the vast majority of us have floating in our heads.

When considering whether to pause or delete a keyword that has no conversions after a certain amount of traffic, our common sense can inform that judgement, but our intuition is likely stronger on the qualitative aspects of that decision (“There’s no obvious difference between this keyword and a dozen others that are converting as expected.”) than the quantitative aspects.

Achieving a clearer signal with more data

Now consider the flip side of the previous scenario: if we have a keyword with a true conversion rate of 2 percent, how many clicks will it take before the probability of that keyword producing zero conversions falls below 1 percent? The math works out to 228 clicks.

That’s not even the heavy lifting of paid search bidding, where we need to set bids that accurately reflect the underlying conversion rate of a keyword, not just rule out extreme possibilities.

Giving that 2 percent conversion rate keyword 500 clicks to do its job, we’d be right to assume that, on average, it will generate 10 conversions. But the probability of getting exactly 10 conversions is a little under 13 percent. Just one more conversion or less and our observed conversion rate will be 10 percent different from the true conversion rate (running at either 1.8 percent or less, or 2.2 percent or more).

In other words, if we are bidding a keyword with a true conversion rate of 2 percent to a cost per conversion or cost per acquisition target, there is an 87 percent chance that our bid will be off by at least 10 percent if we have 500 clicks’ worth of data. That probability sounds high, but it turns out you need a really large set of data before a keyword’s observed conversion rate will consistently mirror its true conversion rate.

Staying with the same example, if you wanted to reduce the chance of your bids being off by 10 percent or more to a probability of less than 10 percent, you would need over 13,500 clicks for a keyword with a true conversion rate of 2 percent. That’s just not practical, or even possible, for a great many search programs and their keywords.

This raises two related questions that are fundamental to how a paid search program is bid and managed:

    How aggressive do we want to be in setting individual keyword bids?How are we going to aggregate data across keywords to set more accurate bids for each keyword individually?

To set a more accurate bid for an individual keyword, you can essentially wait until it has accumulated more data and/or use data from other keywords to inform its bid. Being “aggressive” in setting an individual keyword’s bid would be favoring using that keyword’s own data even when the error bars on estimating its conversion rate are fairly wide.

A more aggressive approach supposes that some keywords will inherently perform differently from even their closest keyword “cousins,” so it will ultimately be beneficial to more quickly limit the influence that results from related keywords have on individual keyword bids.

For example, one of the simplest (and probably still most common) ways that a paid search advertiser can deal with sparse individual keyword data is to aggregate data at the ad group level or up to the campaign or even account level. The ad group may generate a one percent conversion rate overall, but the advertiser believes that the true conversion rate of the individual keywords varies a great deal.

By bidding keywords completely by their own individual data when they have achieved 500 or 1,000 clicks, the advertiser knows that statistical chance will lead to bids that are off by 50 percent or more at any given time for a non-trivial share of the keywords achieving that level of volume, but that may be worth it.

For a keyword with a true conversion rate of 2 percent, observed conversion rate will differ by plus or minus 50 percent from the true conversion rate about 15 percent of the time, on average, after 500 clicks, and 3 percent of the time after 1,000 clicks. If the alternative is for that keyword to get its bid from the ad group (based on its one percent conversion rate), then that will still be better than having a bid that is 50 percent too low 100 percent of the time.

This speaks to the importance of wisely grouping keywords together for bidding purposes. For an advertiser whose bidding platform is confined to using the hierarchical structure of their AdWords paid search account to aggregate data, this means creating ad groups of keywords that are likely to convert very similarly.

Often this will happen naturally, but not always, and there are more sophisticated ways to aggregate data across keywords if we don’t have to confine our thinking to the traditional ad group/campaign/account model.

Predicting conversion rate based on keyword attributes

There is a lot we can know about an individual keyword and the attributes it shares with keywords that we may or may not want to group in the same ad group or campaign for any number of reasons (ad copy, audience targeting, location targeting and so on)

The number of keyword attributes that could be meaningful in predicting conversion rates is limited only by an advertiser’s imagination, but some examples include attributes of the products or services the keyword is promoting:

product category and subcategories;landing page;color;size;material;gender;price range;promotional status;manufacturer and so on.

We can also consider aspects of the keyword itself, like whether it contains a manufacturer name or model number; the individual words or “tokens” it contains (like “cheap” vs. “designer”); whether it contains the advertiser’s brand name; its match type; its character length and on and on.

Not all attributes of a keyword we can think of will be great predictors of conversion performance or even generate enough volume for us to do a useful analysis, but approaching bidding in this way opens up our possibilities in dealing with the problem of thin data at the individual keyword level. Google itself has dabbled in this line of thinking with AdWords labels, though it has its limits.

When considering multiple keyword attributes in paid search bidding, the level of mathematical complexity can escalate very quickly, but even approaches on the simpler end of the spectrum can be effective at producing more accurate keyword bidding decisions.

Closing thoughts

I’ve really just scratched the surface on the topic of predicting keyword conversion rates and the basic statistics that surround paid search bidding. Most advertisers also have to consider some form of average order size or value, and seasonality can have a huge effect on where we want our bids to be.

Paid search bidding has also only grown more complex over time as properly accounting for factors like device, audience and geography have grown more important.

Clearly, there are many moving pieces here, and while our intuition may not always be sound when scanning through monthly keyword-level performance results, we can trust it a bit more in assessing whether the tools we are using to help us make better decisions are actually doing so smartly and delivering the kind of higher-level results that meet our expectations over the long term.

Opinions expressed in this article are those of the guest author and not necessarily Search Engine Land. Staff authors are listed here.

The ins and outs of keyword research

A great way to expand your PPC account is by performing keyword research in order to uncover untapped opportunities. Keyword research is a key tactic for growing an account — especially new accounts, but even mature accounts can stand to benefit from ongoing expansions.

When keyword research does & does not make sense

Keyword research is sometimes debated as being unnecessary, and it is fair to say that certain methods of keyword research are futile. Back in the day, it wasn’t uncommon to hear search marketers taking pride in the number of keywords managed, but it’s a different ballgame now; keyword counts don’t correlate to positive performance or even account health. All that to say, there’s no need to add keywords just for the sake of adding keywords!

A good goal with keyword research is to try to identify entirely new themes. There may be some room to flesh out existing themes, but there’s only so much expansion on existing keywords before you’ll pass the point of diminishing returns due to lack of volume as keywords become more and more specific or similar to existing terms.

For example, instead of trying to find every possible hyper-specific description of a shirt (“pink polka dotted yellow shirt”), look for other opportunities to branch out, like tapping into “tops” terms, or terms for different types of shirts, like “blouses,” “sweaters,” “tees,” “button-ups,” “sweat shirts,” “fleeces” and “cardigans.”

You can also look for additional qualities or benefits that might have enough volume to be worth expanding into. You may have already exhausted terms to describe warm sweatshirts, but there are other qualities that consumers search for, such as “half-zip,” “pull-over,” or “hoodie.” Identifying new themes for root terms and adjectives will help you to uncover worthwhile keyword expansions.

Beyond that, it helps to check out the keyword search volume estimates to see if keywords are even worth the time. There’s no use cluttering your account with terms that aren’t going to have a real impact — not to mention the waste of time that it would take to organize said keywords into campaigns.

The same can be said for keywords that are really similar to, or possibly even already covered by, keywords that are in your account already. It may make sense on a one-off basis to add keywords of this nature if you intend to set a different bid and/or if you believe that search query matching will make it worthwhile to have both terms.

Be careful about adding these types of keywords in bulk without a plan to keep an eye on performance. Additions of keywords that are highly similar to existing terms should be monitored closely to ensure that the new keywords contribute to incremental gains, as opposed to robbing Peter to pay Paul.

Monitoring the impact of new keywords

In order to monitor performance to ensure that the new keywords are adding value to the account, I highly recommend labeling the new keywords as they are added. Labeling the new keywords allows for quick and easy performance reviews. Furthermore, it is good practice to analyze positive performance, as well as poor performance. Make certain that positive performance of new keywords isn’t to the detriment of existing keywords — or, if it is, then make certain that overall account performance is stronger.

In addition to monitoring keyword performance, I also prefer to get rid of keywords that are flagged for low search volume. Now that keywords can truly be deleted if they haven’t generated impressions, I prefer to get rid of these keywords so that they aren’t cluttering the account. In large accounts, downloading and syncing the account can take quite a bit of time, so there’s no need to further bog it down with unnecessary keywords.

Beware the broad match duplicate keywords

Most advertisers are pretty careful to avoid duplicate keywords to ensure that the same keyword isn’t competing against itself. However, when auditing accounts, I have often found broad match keywords that compete against each other.

Since broad match keywords don’t rely on the order of the terms, two different keywords can still be considered duplicates. For example, the terms yellow rain jacket and rain jacket yellow would compete for the same queries.

Tips for identifying worthwhile keywords

Expand upon your top performers. Take a look at your best performing keywords and search terms to brainstorm additional variations, themes or qualifiers that might bring in incremental volume. Be wary of adding terms that are too close to existing terms — for the reasons given above — but your top performers may help you identify qualifiers that could be used with other root terms. Putting these terms into a keyword research tool may help to cover other relevant terms, as well.Talk the talk. In order to manage an account well, it’s imperative to understand the business and the industry. These means understanding industry terms and the features, benefits and differentiators. All of these qualifiers can help drive keyword research and open up new themes.Think like a consumer. There are industry terms and then there are layman’s terms. Sometimes advertisers get so caught up in their own professional terminology that they don’t account for the ways that someone without technical knowledge might seek out their products. This is especially true of B2C. Of course the exception makes the rule, and there are certain instances where it makes sense to maintain an account with only highly specific, technical keywords but generally speaking, there is more volume and therefore more profit (even if at a slightly lower ROAS) in layman’s terms.Identify gaps in organic traffic. Monitoring Google Analytics organic traffic trends can help to determine areas where organic traffic has decreased, or where there isn’t much traction. If you aren’t already bidding on said terms, you might consider adding them to your paid search accounts to become more aggressive.Competitive research. Where better to identify industry terms than to look toward your competitors? There are various tools that you can use to look at terms your competitors are bidding on, which can then be compared to your own keyword list to identify opportunities. Don’t limit yourself toward only using paid search tools – you can also use SEO tools to identify keywords based upon analysis of your competitors’ websites, as well.

Opinions expressed in this article are those of the guest author and not necessarily Search Engine Land. Staff authors are listed here.

Integrate lifetime value to benefit a customer-led SEO strategy

I recently attended Edinburgh’s Turing Festival. (It’s a fantastic tech conference, and of course, Edinburgh is a fantastic city; you should go!)

Speakers included many names familiar to those in the online marketing industry, such as Moz’s Rand Fishkin, Seer Interactive’s Wil Reynolds and Unbounce’s Oli Gardner. And while each presentation had a high standard with thoughtful takeaways, the most insightful speaker to me was Andy Young.

Young spoke about data-driven growth, and although his presentation was more tailored to startups, it had strong parallels for in-house and agency marketers looking to help their businesses or clients gain an edge in their digital marketing strategy. In particular, he talked about how to build cohort segmentation into your web performance analysis to discover which marketing channels were generating better customers.

Young neatly defined solving lifetime value analysis as a subset of correctly attributing direct traffic, so that a user can be further classified as part of an LTV assessment at a later date. (Another challenge on our path will be assessing our attribution model generally, but let’s leave that for another day.)

Here are some key approaches:

Appropriately tag visitors to common “existing customer” URLs. For example, visitors to or are users with an existing account that can be matched back to their original converting UTM session. This will likely allow you to tag up to 50 percent of your direct traffic for B2C businesses.Ensure app deep link tracking feeds back into your data warehouse (for example, Branch, Adjust, Appsflyer).Offline, ensure customers are asked where they came from and/or how they first heard about you — and similarly attribute back.Use call tracking. You can if you run paid campaigns. Do it!Use landing pages or custom tooling in your web analytics platform to tag free social, email, product notifications and so on, so you’re putting customers into the funnel as cleanly attributed as possible.

I strongly agree with Andy Young here that it is usually possible to get everything tracked — what is often lacking is the will. The main takeaway from Young’s talk that you need to fix your analytics gaps.

Lifetime value and cohort behaviors

Taking the baton, what I find most useful as an outcome of this streamlined thinking is that it solves the majority of gaps for creating a LTV (or Lifetime Value) calculation that adds a lot of value for organic strategy in particular.

At the very least, you can get to a pretty good approximation of a LTV calculation for the majority of your customers and get a useful insight and clear action out of your calculations to test back against its impact on your business bottom line.

If you understand your customer cohort behaviors, you can work to a LTV that allows you to test progress along the expected cohort path to their ultimate LTV. So you don’t need to wait six or 12 months before you declare your LTV model a success or off the mark — you can implement it and check if your cohorts are going in the right direction.

If they are, then you are already winning. If you’ve done a really good job of tagging, you can probably refine further at this early stage, the better to drive competitive advantage for your clients.

Why is it useful to know your LTV? To keep the answer simple, if your CoS (Cost of Sale) in a particular channel is lower than your LTV, then you have a business. More importantly for digital marketers — if you don’t know your LTV, it’s quite likely you may be…

Underspending on your real maximum CPCs in paid search.Not spending enough in front of the right demographics in programmatic or classic display, or via email promotion.Not offering perks (such as, say, free delivery at lower spends) which have serious impact on your conversion rates.Not prioritizing page load performance enough. (I mean, seriously, it’s getting worse.)

How does this relate to SEO strategy?

I take a view of organic traffic growth as a path to organic customer growth. That path needs to work with your other marketing channels — particularly paid search, as you are operating on the same playing field (the SERP) and showing a marketing message to the same eyeballs.

You can’t correctly prioritize target organic keyphrase terms if you don’t know how much they contribute to your high-LTV customers. And similarly, if you have to reconfigure your paid campaigns to better prioritize terms a particular high-value cohort has, you need to support that shortfall with organic search.

In short: Your overarching keyphrase strategy should be matched to cohort analysis resting on a LTV value model.

In the same vein, page call-to-action (CTA) messaging value can be assessed by looking at your organic and paid cohort’s contribution to LTV further down the funnel: Do you get a higher initial spend but lower LTV customer with CTA A (20-percent discount and free delivery) versus CTA B (Tell a friend and we’ll give you a $50 gift voucher) in the SERPs? Is one CTA in particular better for organic last-click converters or paid? Do brand term searches on second devices (mobile/tablet at home that evening) for that UTM go up with a combined message across each channel for the initial generic term?

So, in short, to get to a sophisticated keyphrase strategy that targets your most valuable customers over time, make sure you do one thing first: get tagging!

Opinions expressed in this article are those of the guest author and not necessarily Search Engine Land. Staff authors are listed here.

How to get the old Google Keyword Planner volume data

Back in June, the AdWords Keyword Planner started displaying the same volume of searches for keywords that were similar. Search Engine Land reported on the anomalies and asked Google about it but didn’t receive a response at the time. Ginny Marvin speculated, and now an AdWords representative has confirmed, that this was caused by close variants.

@tanestis Hi there, terms that are close variants of each other will show the same volumes within the Keyword Planner. -Mitch

— Google AdWords (@adwords) July 21, 2016

Many people and tools are overestimating search volume by adding together the results that AdWords Keyword Planner has already grouped, effectively doubling or tripling volume estimates.

If you’re like me, you want to see more data, not less. You want to be able to determine which version of a keyword you want to target. You know that even small changes to a word can mean huge changes to intent and value.

How do we get our data back?

Using Ginny’s example of “cat” and “cats,” we can see that Google Keyword Planner shows us average monthly searches of 3,350,000 for each. Add these to the plan and review plan, or, alternatively, just start with “Plan your budget and get forecasts” instead of “Find new keywords and get search volume data.”

[Click to enlarge]

When you are reviewing the plan, you need to change the date range, enter a bid and click on the keyword tab, as shown below. You can also adjust the match type to your preference.

[Click to enlarge]

You will now have impression volume for the individual keywords. Note that this may include occasions when you would also be shown on page 2 of the search results. The caveat to all of this is that you have to know the different variations that you want to see the data for, and I don’t believe there’s a good way to get all of them.

Why should you care?

Simply put, not all words are created equal. They don’t have the same intent or the same value. A recent example from work would be the differences for “api test,” “api tester” and “api testing.” Even AdWords shows vast differences in the search volumes and cost per click.

Unless, of course, you’re relying on Keyword Planner now…

On a side note, check out Daniel Gilbert’s AdWords script that gets rid of close variants. The fact that this even exists should tell Google that this change was a bad idea. It was optional before it was forced on paid search marketers, and now it’s being forced on those doing research for organic as well.

Opinions expressed in this article are those of the guest author and not necessarily Search Engine Land. Staff authors are listed here.

Machine learning for large-scale SEM accounts

A key challenge when working on what we could term “large-scale” PPC accounts is efficiency. There will always be more that you could do if given an unlimited amount of time to build out and optimize an AdWords campaign; therefore, the trick is managing priorities and being efficient with your time.

In this post, I will talk about how concepts from machine learning could potentially be applied to help with the efficiency part. I’ll use keyword categorization as an example.

To paraphrase Steve Jobs, a computer is like “a bicycle for the mind.” The generally understood meaning of this statement is that, in the same way a bike can increase the efficiency of human-powered locomotion, computers can increase human mental productivity and output.

With the existentialism out of the way, let’s get to something tangible — We’ll explore here how relevant/valuable it could be to try and automate the process of placing new key phrases into an existing campaign.

What do we mean by machine learning?

As a definition of sorts that ties into our objectives, let’s consider the following to be true:

Machine learning is a method used to devise models/algorithms that allow prediction. These models allow users to produce reliable, repeatable decisions and results by learning from historical relationships and trends in data.

The benefit of “reliable, repeatable decisions” is the exact value that we’re interested in achieving in this case.

At a very high level, the objective of a machine learning algorithm is to output a prediction formula and related coefficients which it has found to minimize the the “error” – i.e., has been rigorously found to have the greatest predictive power.

The two main types of problems solved by machine learning applications are classification and regression. Classification relates to predicting which label should be applied to data, whereas regression predicts a continuous variable (the simplest example being taking a line-of-best-fit).

Categorization of keywords as a “classification” problem

With this in mind, my goal is to show how text classification could be used to programmatically decide where newly surfaced key phrases (e.g., from regular search query reports) can be placed. This is a trivial exercise, but an important one which can be time-consuming to keep on top of when you have an account of any scale.

A primary prerequisite for solving a classification problem is some already classified data. Given that an existing paid search account has keywords “classified” by the campaign they are in, this is a good place to begin.

The next requirement is some “features” that can be used to try and predict what the classification of new data should be. “Features” are essentially the elements on which a model is built — the predictor variables.

The simplest way to transform text data into a feature which is useful to an algorithm is to create a “bag of words” vector. This is simply a vector which contains a count of the number of times a word exists in a given document. In our case, we’re treating a keyword as a very, very short document.

Note: In practice, because our “documents” –i.e., keywords — are short, we could end up with a set of vectors that is not meaningful enough due to lack of diversity, but it’s out of the scope of this article to delve further into this.

Selecting a suitable algorithm

There is a wide range of different algorithmic approaches to solving a wide range of problem types. The below image illustrates this and also shows that there is a certain underlying logic which can help direct you toward a suitable choice.


As we’re in the area of text classification, let’s implement a Naive Bayes model to see if there is potential in this approach. It’s a pretty simple model (hence the “naive” part), but the fact that our feature set is pretty simple means it could actually pair up quite well with that.

I won’t go into any detail of how to apply this model, other than sharing how I would implement this in Python using the scikit-learn package — the reason being that I want to illustrate that it’s possible to leverage the power of machine learning’s predictive capabilities on only a few lines of code.*

Below is a fairly exciting screen shot of my Jupyter Notebook going through the key steps of:

loading the data with which to build the model (~20,000 keyphrases, pre-classified);splitting my data into training and testing subsets (This is necessary so we can “test” that our model will actually predict on future data and not just describe historical data);creating a basic pipeline which a) creates the features as discussed (CountVectorizer) and b) applies the selected method (MultinomialNB); andpredicting values for the “test” set and gauging how accurate the labeling is vs. the “true” values.

*The caveat here is that I’m not a developer, but a mathematician who programs as a means to a mathematical end.


So, how effective is this? Using a simple measure of accuracy, this method correctly labeled/categorized 91 percent of “new” key phrases (4,431 out of 4,869).

Whilst this could be considered a decent result, we’d do a lot more tuning and testing before putting a model like this into practice.

However, I do believe it provides sufficient evidence that this is a relevant approach that can be taken forward to improve and automate processes — thus achieving the objective of gaining efficiency at scale via reliable, repeatable decision.

Opinions expressed in this article are those of the guest author and not necessarily Search Engine Land. Staff authors are listed here.

Broad match: What is it good for?

Walk into any bar (or type your way into any digital space) where PPC people are gathered, and you will find a thousand different methods for using the different match types available when adding keywords into your campaigns.

None of them are inherently wrong — ultimately, it’s the results that matter. But is there a method that can make effective use of broad match keywords without ruining our chances at a healthy cost per conversion? That’s the subject this column will cover, while also delving heavily into the area of keyword research, so get ready to have some fun.

Broad match = hunter-gatherer

We often use broad match keywords when starting paid search campaigns from scratch, as we may not intuitively know more specific searches that people might be using. Other times, we encounter a large volume of broad match keywords when we take over management on an AdWords account set up only using broad match (It happens!).

A typical next step in either of these situations would be to duplicate these keywords into exact and phrase match, then monitor them for new keyword opportunities. We might even supplement this with research from a tool like SpyFu to pull competitor keywords.

The weakness of broad match keywords is that they can be triggered by literally anything even remotely close to the keyword; when you added the broad match keyword “hammock” to your newest campaign, you probably didn’t intend for the search “banana hammock” to trigger it and show your ad. These are the sorts of hijinks that can happen when running a PPC campaign.

By now, most AdWords users know that the Search Terms report gathers search information and allows us to add keywords, or add negative keywords, from this. But using the Search Terms report, we can turn the weakness of broad match keywords into the best keyword research tool available: actual search data.

Broad match keywords are the new keyword research

Struggling with the keyword research process on a new campaign? Just starting advertising for a new client who’s never done AdWords, and the keyword planner isn’t being useful? Try building AdWords campaigns composed of broad match keywords, with very minimal bids (targeting 5 to 9 ad position, or even lower), and use the Search Terms report to add more specific keywords that come directly from the users.

Using the Search Terms report in this manner turns your keywords into a research tool, giving you real data to use when adding long-tail searches as exact match or phrase match keywords. Doing this will effectively let you build very specific and detailed campaigns at very little cost.

Bidding strategy?

Beyond using broad match keywords with minimal bidding, how can we work this thought process into our ongoing campaigns side-by-side with exact and phrase match keywords? This can be tough. Often, we see some really great, consistent results from broad match keywords that were triggered by highly relevant searches, but we must consider the wide variance possible with these searches and adapt our bid strategies accordingly.

When campaigns are running, we cannot simply set bids at $0.10 for our broad match keywords anymore and look for great keywords. Instead, consider the following.

Broad match: Bid for position 3 or 4.Phrase match: Semi-aggressive, bid for positions 1 to 3, depending on cost.Exact match: Aggressive, bid for top of page.

In the above example, which shows variations of the keyword “hammock,” we would also be sure to include [banana hammock] as a negative keyword. We might use this match type-based bid strategy because the variance of a broad match keyword makes the searches that trigger it entirely too unpredictable, but we still need to be in a position to convert searchers with our broad match keywords in the hopes of identifying the searches that we must be going after.

These search terms that convert from broad match should be added as phrase/exact match with higher bids and become the focal point of our campaigns; we know with a better degree of certainty what people are searching for here and should therefore be willing to pay more for their clicks.

Following this process should eliminate the potential trial-and-error process you run into with adding untested keywords into campaigns and adjusting bids based on performance without a strategic base. Consider this as a base to start revising your campaigns; from here, let the data tell you where to go.

Opinions expressed in this article are those of the guest author and not necessarily Search Engine Land. Staff authors are listed here.