AdWords is a beast.
First off, it’s a moving target. In the last year alone, Google rolled out over one thousand changes to AdWords — that’s twenty per week. New ways to target, new reports, new UIs.
On top of this, the channel becomes a more efficient by the hour, and by ‘efficient’ I mean “keywords priced such that your return goes to zero.”
Raise your hands if your bids have increased over time.
When I started driving AdWords traffic to my retailer back in 2003, I was paying $.05 per click. Today some of those same clicks cost over $200 — about what we’d make on an initial sale.
Add in the challenges that are unique to ecommerce — lower margins and conversion rates (than lead generation businesses, for example) — and you have a deck stacked against you. This is why most businesses today lose money on AdWords.
That’s right — most retailers are losing money on the channel.
But this doesn’t have to be the case. AdWords is a brilliant channel. No other marketing channel promises to bring you relevant buyers searching for your products within hours. At my company, AdWords got us our first order and typically contributed ten to twenty percent of our revenue each year. And the sky really is the limit: I’ve worked with retailers who employ only AdWords and have grown north of $100M in revenue because of it.
One hundred million dollars in revenue off of AdWords. Want to know how?
In this post I lay out a killer AdWords approach that works. With a hat-tip to the Seattle Seahawks let’s call it AdWords Beast Mode(Lynch, would be honored . . . right? I actually did research this and he did not extend his trademarked phrase to AdWords. And going out on a limb here: he probably never will.). It’s an overview of the approach we use successfully at Mineral.io with all of our paid search clients.
Want to download this article for later? Click here to download it as a PDF. Includes my AdWords “Beast Mode” checklist (discussed later in this article).
AdWords Beast Mode entails how you should think about paid search, how you should structure your campaigns, and how you should measure your success.
It takes time, but it’s worth it: if you aren’t making money on AdWords, you should be. If you are, you could be making more.
To understand why AdWords Beast Mode works, we need to delve a bit deeper on why most businesses lose money on AdWords.
Yes, bidding is expensive and ecommerce margins don’t give you a whole lot of wiggle room.
But the more fundamental reasons most businesses don’t succeed are that they 1) don’t track the right things and 2) choose Set It and Forget It as a strategy.
Regarding the former, most businesses I’ve seen have no idea how they are doing on AdWords. They might fixate on vanity metrics like their Average Position or Quality Score and not on their bottom-line return or on the quality of the customers they are driving to their site. AdWords might look like it is working — ads are in top spots and driving a heaps of traffic after all — but the ads are driving undesirable customers that return everything they buy and drain the life force from a customer service department.
Regarding the latter, these same businesses often choose Set It and Forget It as a strategy. That’s right, they select keyword phrases, build out their Ad Groups, choose bids that get them a desired position, then move on to more sexy things like content marketing. These businesses ignore the fact that AdWords is a multiplayer game. Think World of Warcraft, not Pong. Your actions impact competitors’ actions. You change your bid, they change theirs — along with their ad copy, landing pages, CTRs, etc.
As AdWords legend Dave Collins likes to say, “AdWords is the most confrontational form of advertising in existence.” In this environment, your competitors’ gain is your loss. Make no mistake: they are out to get you, and there are a lot of them — hundreds, even thousands per keyword. Don’t believe me? Play with some of these tools to see for yourself. An Ad Group that gave you a positive ROI when you created it may quickly go negative — and stay negative because you aren’t measuring ROI in the first place.
Losses ensue. Enter the Beast.
The first step towards going Beast Mode on AdWords is to hit the pause button, step back, and think about AdWords as a holistic marketing process flow. What do I mean by ‘flow’? I mean you have to think not in terms of keywords or landing pages or ads, but of all of them combined in sequence.
Memorize this flow:
AdWords Beast Mode Process Flow:
Precise Persona ⇒ Precise Keywords ⇒ Precise Ads ⇒ Precise Landing Pages ⇒ Sales!
A holistic marketing process flow means starting with a customer persona in mind. Then work through what keywords that customer would be searching for, what ads would align with her intent, and what landing pages would compel her to take action.
Sounds simple, I know, but nobody does it. Not a lot of big retailers, not a lot of paid search agencies. The more common, default approach is to use a keyword tool to generate a ton of target keywords, unthinkingly draft some ads that direct traffic for those keywords to existing pages on your site, and use a bidding tool to optimize around a certain position. Then, if you want more traffic, up your bids.
This default approach is close, but no cigar.
Two paid search techniques will drive more traffic to your site: increasing your bids and increasing your relevance. And of the two, most retailers stubbornly choose the former even though Google consistently rewards the latter:
Relevant ads tend to earn more clicks, appear in a higher position, and bring you the most success, so it’s important to focus on relevance when creating your campaigns. Having relevant ads could also decrease your cost-per-click, saving you money with each click. Non-relevant ads tend to earn fewer clicks, or they might not even show up at all.
fricking Google, ok?
To reduce cost and increase conversions, increase relevancy across your Flow. Let me elaborate.
AdWords Beast Mode all starts with targeting the right customer persona. No, not all customers are created equal. Some are low lifetime value (LTV) undesirables: they buy-and-return, or buy low-margin products, or buy once and never again. Others are high LTV gems: they buy repeatedly or buy high margin items or never require any customer support.
Focus on the gems. They drive your business. (I’ll lay out specific methods to identify them in future posts, but for now just sort your customer database by frequency of purchase and go with the top ones.) Then answer some of these questions to develop one or two customer personas:
What do your gems have in common? Are they male, female? Do they live in certain cities?
What are they buying? High LTV customers often buy from one or two categories more than others.
Why are they buying what they are buying? What problem are they trying to solve? This is critical because it gets at purchase intent. Don’t guess — ask them or ask your customer service team.
With a persona (and her motivations) in mind, now you can break out your favorite keyword tool and move on to choosing your keywords.
Choosing precise keywords requires some basic analytical skills. Think outside of your brand and product names to discover which keywords your target persona would search for given her motivation.
Ideally, these keywords should be 2-3 words long and get at the intent behind the search. Here’s a great example for a diet shake that contains L-Glutamine (an amino acid that makes you feel full and reduces cravings):
Bad: “l-glutamine shake” (too product-focused, how does your persona know what L-Glutamine is in the first place?)
Better: “weight-loss shake” (good, but does the persona even know she needs a shake?)
Best: “lose weight for spring break” (great, gets at intent with a deadline)
Now, before you get all hot and bothered, I do think that bidding on specific terms germane to the solution in question like “L-Glutamine Shake” would work in the short term — the web shopper in question knows exactly what he or she wants and is therefore likely to convert. But these are also the terms that are 1) too narrow to drive sufficient traffic over the long-term and 2) apt to get bid up quickly until they become marginally profitable for your business. By all means add them, but for AdWords Beast Mode to work we want to cast a wider net and include terms that get at intent.
So research your persona ad infinitum. Use your website’s search query report, AdWords own Keyword Tool, and the Search Term Report to find what people search for, and keep in mind that you can never have too many phrases. This bears repeating: don’t ever stop generating keywords, you’ll be surprised what drives traffic.
Sidebar on AdWords match types: of all the match types, I’d recommend staying away from straight broad match as it tends to attract higher cost, less targeted traffic. Exact match will have a higher CPC, but it is a better converter, especially when combined with the techniques I describe in this article. And use negative keywords like a boss to eliminate related-but-useless terms like “free”.
Above all, get at intent with your keywords.
Too much to read in one sitting? Click here to download this article as a PDF. Includes my AdWords “Beast Mode” checklist (discussed later in this article).
Creating precise ads requires some basic copywriting skills. You’ll need to use empathy to capture the needs, wants, and intentions that are driving your ideal persona’s search.
Let’s explore this in the men’s business suit category. Think of all the personas there. Some men want suits of a certain style or color. Other men want suits that travel well. And a few jackasses want the classic orange-and-blue Dumb and Dumber tuxedos for Halloween.
Say I run a men’s apparel retailer that has identified those who want first interview suits as our target persona. They want a solid, safe suit that makes them look great, and after they get the job they’ll buy all their future business attire from my company. Hence the high LTV.
These personas might use the phrase get business interview suit when they hit up Google.
If I were running paid search for this retailer, I’d want my ad copy to reflect the persona’s intent, leading with verbiage like “Can’t-fail Interview Suits” or “Get the Suit, Get the Job”. This copy speaks to the urgency and importance of the occasion, the primary intent behind our persona when he searches for get business interview suit.
But observe the actual results:
In the top position, we have Men’s Wearhouse. They are bidding an arm and a leg for their top spot — not surprising for a large retailer. But their ad has little relevance to our persona — it mentions “Top Brands” (what first-time suit buyer knows or cares about suit brands?) and irrelevant categories like “Coats and Outerwear”.
Ladies and gentlemen, I give you your typical AdWords competitor: big budget, fixated on position, willing to blindly pay more in an effort to drive (expensive / unqualified) traffic.
But Esuit.com is in the second position with a much more relevant ad, complete with a custom URL and copy that speaks to the motivation of our persona. I am willing to bet that they are paying considerably less for their #2 spot than that position would suggest — that Google is rewarding their relevancy with a higher spot for a lower bid amount.
This is precise ad copy. Use it and you can roll with the big boys. Not only roll with them, but probably trounce them on metrics like click-thru-rate (CTR) and cost-per-click (CPC).
And don’t stop there — even if you think you’ve nailed it, always create two ads for every keyword and destination URL so you can A/B test and keep improving your results.
Precise ad copy, informed by our persona’s intent, is enticingly effective at driving cheaper clicks.
So the AdWords Beast Mode approach has gotten us a big, fat, juicy high-LTV customer clicking our ad at a fraction of the cost of our competition. Now we have to convert her which is where precise landing pages come in.
Here again, Google rewards relevancy. Make sure you’re following best practices for ecommerce landing pages by linking to a relevant, specific page within your website. Never use your homepage as a landing page, because your users likely won’t take the initiative to find the right page. They’ll simply click the back button to click on another ad.
Instead use your category or product pages as landing pages. Better yet, AB test them against even more relevant, custom pages using a service like Unbounce.
Whatever you do, just make sure that your landing pages are relevant. Let’s take another example of a female persona searching for a black cocktail dress:
Here, it looks like Macy’s copy nails the precise ad copy part (i.e., the copy mentions the search terms), whereas ModCloth’s ad copy leaves a bit to be desired (no mention of black). (Ideally, I’d like to try ad copy that speaks to an intent of “looking amazing” in a black cocktail dress.). So I’m guessing Macy’s deservedly gets considerably more clicks given their better relevance (i.e., they achieve a higher CTR).
But click through to the landing pages, and you’ll see a different story. First Macy’s:
Whoops! The main section header is targeting bridesmaids. How many wedding parties wear black cocktail dresses? Even I know that they answer is “none” unless it’s a Goth wedding. Also, a scroll down the landing page reveals dresses of all colors for all occasions. The onus is on our user, with limited time, to figure out the damn filtering to the left in order to find her black dress.
This isn’t going to happen, so it’s a landing page fail for Macy’s.
Contrast Macy’s landing page with Modcloth’s:
This is a step in the right direction, as I only see black cocktail dresses. Ideally, I’d probably want the main callout to speak to that category, which it does not, but Modcloth does pwn Macy’s on landing page relevancy.
(What I find surprising is that the same Modcloth dude who built out the landing page for this term obviously knows what our persona is looking for, so why did he or she leave clicks on the table by not optimizing the ad copy in the first place?)
Macy’s is doing a better job driving clicks (high CTR), but the clicking shoppers are less likely to convert (lower conversion rate). The result is higher AdWords costs. And Modcloth is doing a better job converting, but could be driving more clicks with a better ad. The result is a missed opportunity.
Knowing this, if I were an upstart competitor using the holistic AdWords Beast Mode approach across both ad copy and landing pages (and assuming I offer a decent selection of black cocktail dresses), I could have it all: lower costs and higher revenue. I’d be opening a can of whoop-ass on the black cocktail dresses category, able to beat these two retail giants through the power of relevancy on even a very competitive phrase.
I have seen the Beast, and it is me.
I hope I’ve shown you why most businesses lose money on AdWords, but it doesn’t have to be that way. Opportunities exist even in super-competitive categories by exploiting relevance, and this holistic, empathetic approach can get you there.
So why don’t all businesses employ some version of AdWords Beast Mode?
That’s easy — it takes time. Lots of it. At Mineral.io, we spend days setting up campaigns for our clients and hours each week revisiting and optimizing them.
You can’t just plug your product feed into a bidding algorithm and expect to take it to the bank. But therein lies the opportunity — most retailers do just that, content to have “checked the AdWords box.”
And this is why you can beat them.
Given the time it takes to employ this strategy, you may be wondering…where should you start?
I’d recommend walking before you run. Use Pareto’s Principle and start with a select few ad campaigns targeting five to ten of your top-selling, high margin products or categories.
Then follow my AdWords Beast Mode Checklist – a product-first approach to AdWords strategy that will increase the likelihood taht you will achieve the holy grail of lower CPC rates with higher conversion rates.
The checklist is available for free download, and includes my best pro-tips on AdWords and a PDF version of this entire article.
Thinking through, creating, and optimizing campaigns is tedious work, but it’s worth it: you will generate more profits than you would by just upping all your bids across the board. You will also be able to compete successfully against bigger competitors with bigger budgets. Why? Because you are making your marketing more relevant, and Google will reward that relevancy with more clicks at a lower cost.
I hope this overview of our tested AdWords approach works for you. If you want more information on it or how to hack paid search in general, I encourage you to sign up to my email list below. And if you are at all curious about having an agency do this work for you, check out our ecommerce AdWords program over at Mineral.io.