Article

The 7 Components of a Winning Advertising Strategy

Advertising Analytics Consumer Behavior Data google ads Meta

by Sam Tomlinson

February 19, 2024

I’ve had multiple calls with clients and prospects regarding their advertising and paid media strategy, what role it should play in their overall growth, and the critical components that will help them get from where they are today to where they want to be tomorrow. 

My first takeaway from those conversations was the staggering level of misinformation out there about how these platforms work and what brands should be doing on them. 

Today’s issue is dedicated to addressing some of that misinformation. 

Let’s start with a basic premise: successful paid media campaigns come down to doing 7 things brilliantly well: 

1. Data Interconnectivity
2. Financial Consolidation
3. Platform Fundamentals
4. Offer & Angle
5. Creative 
6. Experience
7. Iteration + Evolution

That’s it.

Each one of those has a significant amount of complexity, but let’s break them down: 

Data Interconnectivity:

At its most fundamental level, advertising has always been an exercise in capital allocation and maximizing expected value. Digital platforms simply added exponentially more data into the view of what could be considered when performing those calculations – to the point where the calculations required quickly exceeded human capacity to perform them. 

Digital ad platforms addressed this with automation and machine learning, which can perform probabilistic and computational tasks faster, more accurately and in more sophisticated ways. In practicality, that shift resulted in the rise of smart bidding, the integration of generative LLMs, and the removal of manual control levers. 

This has come with two tradeoffs: (1) when you automate something, you shift control over the automated function to a higher level of abstraction. In the case of digital ad platforms, automation and smart bidding have shifted control from the manual, in-platform levers to the data layer that smart bidding algorithms use to calibrate their bids and budgets and optimizations. 

And (2) when (1) happens, the incremental advantage of in-platform data goes to zero, because now every advertiser has access to supercomputers capable of sifting through reams of data to find the hidden gems. 

So, where’s the advantage today? 

It’s still in the data, but instead of platform data, it’s the advertiser’s data. 

The advertisers who can provide platforms with richer, more robust, more accurate, more timely user and business data will win. Those that continue to rely on platform-level data and website-level conversion tracking will find themselves with deteriorating results. 

It is not an exaggeration to say that the #1 in-account factor that holds back performance is a lack of data interconnectivity. If your business does not have the ability to pass incremental, non-public data to ad platforms, you’re at a disadvantage. If you don’t have the ability to track (at the very least) relevant conversions, you might as well light your money on fire. 

Any in-platform advertising strategy that does not begin with data is destined for failure. The only questions are how long will it take and how much money will be blown in the process. 

Financial Consolidation:

Most brands do too much. Most agencies try to sell too much – too many platforms, too many tactics, too many tests, too many things. And all it does is stunt growth and kill performance.

The story is always the same: a brand finds success with one offer/angle on one platform. They start to see sales / leads increase. And immediately, their CMO / agency / Marketing Director / freelancer / whatever recommends to “roll it out” to other platforms, in a valiant (but utterly pointless) effort to scale. Everyone gets all hyped up, launches some stuff, then watches as efficiency and performance sink like the Titanic. After a few months, we rinse & repeat. 

This isn’t just hypothetical, either. I recently spoke to a potential client that was caught up in this exact situation: they had found some success on paid search, with a budget of $10k to $15k per month. Their success on paid search had resulted in a record-setting two months from a business standpoint. Their new agency was recommending Demand Generation campaigns, YouTube campaigns and new Meta campaigns – all on the same budget ($10k to $15k). My advice was that they weren’t ready and they should focus on scaling paid search to $50k to $100k per month. 

They didn’t listen. And their numbers since diversifying make a dumpster fire look like Miss America. 

Let’s get right to it: if a brand is unable to spend a minimum of $50k per month ($600k per year) on a core platform (Google, Meta, Amazon) at or above reasonable efficiency targets, then that brand isn’t fit to compete on another platform, let alone fit to win. 

The reason for this bold claim is simple mathematics: mastering a channel requires focused, compounding learnings. Compounding requires time, energy, effort and resources. If those things are split across multiple channels, each one gets only a fraction of what you have available. Making things even more complex is platform evolution, which creates a drag on those learnings. If you aren’t able to compound your learnings + skills in a platform at a rate that exceeds the rate of platform evolution, you wind up going backward (which is where the real money-losing happens). 

The key – to paraphrase Warren Buffett – is to not interrupt the compounding by introducing new platforms or campaign types or strategies. Keep all of your focus, time, resources and energy pouring into your core platform until you hit scale. 

It’s easy for agencies to recommend doing lots of stuff – it makes them seem like they are doing more and earning their fee. But more activity does not entail more results. 

I realize this flies in the face of much of the advice that others in the paid media / digital advertising space have uttered over the past years. I realize this may not be popular among other agency types. But here’s the simple reality: each of the three core platforms (Google, Meta, Amazon) has a sufficiently large audience to scale a business well beyond $50M/yr (and likely, well beyond $100M).

The bottom line is simple: until you have one core platform operating at a predictable level of efficiency and scale (>$50k/mo), diversification is a net-negative in 99% of cases. 

Trying to do too much, too soon will cost you more – both in terms of efficiency losses (not just on ad returns, but also lower team efficiency, bifurcated focus, etc.) and opportunity costs (each minute spent on another platform is a minute you can’t spend building your foundational platform) – and bring you less. That is not something you can afford.

Platform Fundamentals:

There are no two ways about this – if you’re going to succeed and scale a core platform, you need to be brilliant at the basics. And the bar for that brilliance has only increased over the past 5 years. 

At a bare minimum, this means nailing: 

1. Account Setup
2. Business Data
3. Bidding Strategy
4. Audience Insights / Audience Research
5. Targeting
6. Exclusions (negative audiences, keywords, etc.)
7. Rules + Automations
8. Testing Strategy + Approach

There are no shortcuts to be had; there’s only the work. 

Keyword + Audience research is among the most tedious things marketers do, and it remains one of the most critical and one of the most strongly correlated with success if it is done on an ongoing basis. Most accounts I audit haven’t done major keyword/audience research since the initial setup – all the while, users have changed, SERPs have changed, industries have changed. If the account hasn’t kept up with those changes, it’s likely that it isn’t performing optimally. 

Creating stop-loss and anomalous performance scripts/rules is about as fun as serving as a pit bull’s human chew toy, but – on days like February 14, 2024 – can be the difference between an account overspending by $10k, $50k or $100k, or only $1k. 

Building integrations between your CRM/CDP and Google Ads or Meta Ads is certainly not easy, but if it’s done well, it can provide a data advantage that improves performance by 20%, 40% or more. 

Each of these things levers up to a single, over-arching philosophy I have for media buying: high floor, high ceiling. 

The high floor is created through exclusions, stop-loss rules, well-calibrated bid targets and automations – these serve to prevent capital allocation with a negative expected value. This can be achieved in multiple ways, from cost/bid caps (Meta) to tCPA / tROAS Portfolio Strategies with Max CPC (Google); what’s important is that it is accomplished. 

The high ceiling is created through a scalable structure and relatively few budget limitations (for instance, I wrote about how daily budgets kill campaign performance here). This allows the account to scale in response to audience demand, unencumbered by a budget constraint. 

Finally, and most crucially, there’s one more element of account structure: an embrace of the simple fact that it is better to not spend than it is to spend poorly. When your account spends poorly, it places a commensurately higher burden on all subsequent spending just to get back to your baseline. That’s a recipe for taking bigger risks and bigger swings, with lower expected values.

The Almighty Offer:

While we’re on the topic of painful lessons, here’s one that most marketers still don’t love: 

Marketing is like training: it can take an average athlete and make him/her good. But it can’t take me and turn me into LeBron James. 

There is NOTHING more critical to the success of a brand than crafting a compelling offer that resonates with your core audience segment(s). 

Your offer is more than a discount. In fact, let’s go one step further: discounting is (almost always) a race to the bottom that is unhealthy to win in the short-run and virtually impossible to sustain in the long-run. 

The offer is comprised of three fundamental elements:

1. Product – the thing you’re selling
2. Value (not just price!) – what it costs vs. what you deliver
3. Competitive Positioning – how what you’re selling + what you deliver compares to what your audience could get elsewhere. 

If you don’t have a compelling offer, it doesn’t matter if your ad was written by David Ogilvy himself. It will not sell.

The third fundamental component – the positioning – is the angle. That’s simply how you connect the offer to the audience. It’s the hook. It’s the little insight that connects your product to your audience’s problem/pain. 

But again, if the offer is bad, the angle will not save it. If the offer is good, the angle can make it great. 

When thinking about your offer, consider: 

• Is it better to offer a single, hero product or a bundle?
• What value does the customer expect at our price point? 
• Discounting alternatives to increase perceived value?
• Choice architecture – can you influence selection through defaults or decoys? 

There’s no single right way to design an offer; but it’s foolish to continue spending on an offer that simply isn’t working. Most brands spend way too much time thinking about their creative, and far too little thinking about their offer. Don’t be like those brands. 

Creative:

For years, marketers have (rightfully & correctly) touted the importance of killer creative. Whether it’s copy or imagery or video, your creative is likely to be – statistically – the first thing your target audience sees from your brand. 

And rarely, if ever, is it the only thing present in that moment. Social feeds are chock-full of captivating content – ads, influencers, videos, trends, you name it. 

From a pure mathematical standpoint, your ad has maybe a second to peak the audience’s curiosity, and 2-3 seasons to earn their attention. That’s a tall task. 

While there are many mistakes brands make on creative, I want to focus on the big three: 

1. Stepwise Testing
2. Lack of Creative Diversity
3. Neglecting Copy

Of these, stepwise testing is probably the most prevalent. We create a single ad, see that it performs pretty well, and then we start creating minor variations of it – we change the headline, tweak the colors, adjust the button style, swap out a photo, etc. 

Then we test all of those ads. 

And most of the time, any initial gain from one variant fades as the sample reaches statistical significance, and we’re left with 10 ads that all performed relatively comparably, and no learnings to show for it. 

The process that leads to this is rooted (I think) in noble intent – we want to do things rigorously and methodically. That’s objectively a good thing! But it is rarely practical to do so in the paid media ecosystem. Doing this results in finding a local maxima: we create the best possible version of this particular creative concept. 

But what gets lost/missed is that the best version of this creative concept (a local maxima) might not even be close to the best possible creative concept for the brand (the global maxima). There’s a reason optimization algorithms are intentionally programmed to test outliers and take random paths – it’s to detect if the maxima is local or global. From a creative standpoint, the only way to do this is to take bigger, bolder swings – you have to test entirely new concepts, not just iterations on the same concept. 

The second, and related, problem is a lack of creative diversity: most brands test variations on a creative theme when they should be testing different themes altogether. 

This happens for a variety of reasons, with the most common being either (a) the siren song of rigorous testing or (b) a desire to comply with the directives of the brand police. 

If you’re not sure if this applies to you, there’s a simple test you can take. Ask yourself: 

• Do all of our ads have a similar look and feel? 
• Do all of our ads conform to a single set of brand guidelines?
• Do they all fall into 1-3 common ad formats (UGC, explainers, review templates, etc.)? 
• Do all of our ads have a similar narrative arc? 

If you answered “yes” (however reluctantly) to any of those, you’re probably falling into this trap. 

Why is this a problem? 

Because ads that all function and look and sound the same all tend to appeal to the same people – and that set of people is only a subset of your total target audience. Ads are – by their nature – somewhat educational. And (as we know), different people learn in different ways: some prefer to hear the content, others prefer to see it in action, still others prefer to do, and many of us require a combination of the three to really “get it”. 

Using only a single ad type is like trying to educate 1,000,000 people by only writing on the blackboard. Sure, it’ll work for some. But for the majority, it’s likely to fall short – not because what you wrote was bad or wrong, but because it did not align with their learning style. 

The same is true for creative. A few months ago, I wrote a post with 19 different creative styles you could test – I’d highly recommend checking it out if you’re curious about what else you could (and should!) try. 

And finally, there’s your actual copy. 

This is one that I simply don’t understand, but I see it all the time: brands invest hundreds of thousands of dollars into shiny new videos, only to run them alongside the same, tired, uninspired copy. The same thing is true in paid search, and the same thing is true on landing pages. 

Most brands don’t realize how insidious this problem is – and even fewer are doing something about it: 

• As LLMs (like ChatGPT-4 and GEMINI) become more prevalent, where do you think they’re going to get their copy from? Your website and your landers! If your copy is as mind-numbingly generic as everyone else’s, you’re going to end up just like them. 
• The same is true for auto-generated assets and optimized text in search/social ads – this comes from your site. If your copy is bad, your ads won’t be good! 
• From an organic search perspective, the content that stands out (according to Google’s own guidelines) is the content that is remarkable, comprehensive, expert, authoritative and trustworthy. Content that looks and sounds just like everyone else’s is hardly going to clear that bar. 

Invest in your copy. Your words matter. 

If you don’t want to be like everyone else, start by not sounding like everyone else. 

Experience:

Everything up to this point has been focused on getting the click; the experience is everything that happens after. 

I’ve made no secret that I think most paid traffic should be directed to dedicated landing pages, for the simple reason that most “evergreen” pages are ill-suited to serve as a continuation of the journey that earned the click in the first place. 

As a simple example, if your ads promoted a free demo offer + highlighted a specific pain point, but your home page or PDP or demo page makes no mention of those things, the user is left confused and skeptical – were they just sold a bill of goods? Does your product/service actually do the thing claimed in the ad? How is this related to everything this brand just told me? 

From a pragmatic perspective, the last two things you want in the mind of your newly-acquired website visitor are doubt and confusion. They will tank your conversion rate faster than Doc Rivers will tank a 3-2 series lead. 

At least half of the ad accounts I audit have a creative-lander mismatch – whether that’s an ad with one CTA and a page with another, or an ad that prompts a user to do something (like request a demo) that isn’t possible on the actual site (for instance, because the brand doesn’t offer demos). 

These are self-inflicted, unforced errors. And they have a staggering cost. Just take a look at the below (anonymized) data: 

The numbers here speak for themselves. And this is by no means an extreme or edge case – in fact, I’d call this a conservative improvement from a poor landing page (like a home page or services page or PDP) to a dedicated, well-designed lander. It’s not unrealistic to see conversion rate improvements from 2.0% to 5.0% or 6.5%, along with AOV increases as well (which aren’t included here). 

While the lander is the primary touchpoint in the experience, it is far from the only one: 

• Form/Checkout
• Thank-You Page
• Post-Purchase/Post-Lead Flow
• Nurture Flow
• Customer Service/Sales Experience

All play a role, too – and every one should be aligned to your angle, offer and creative. If you’re not actively reviewing and auditing your experience, you’re likely not getting the most from your paid media campaigns. 

Iteration & Evolution:

Paid media is an evolve-or-die business. The dynamic nature of digital platforms means that even the best-designed campaigns, creative, landing pages and data flows will, eventually, become obsolete. 

The only way to stay ahead of the curve is to iterate and evolve.

Most platforms make this relatively easy to configure (Meta & Google have made A/B tests and experiments wonderfully simple, if oft-underutilized), but maddeningly difficult to do properly. 

It is no secret that I fundamentally disagree with how most people approach experimentation and testing in paid media, which tends to be a sad middle ground between methodical iteration and radical ideation. 

Fortunately, if you want to do this better, there’s the 10% or 10x Philosophy. 

If you can nail these seven things, you’ll be just fine. If you don’t….you probably won’t. The choice is yours. 

Until next time,

Sam

Article

14 Ways To Leverage Zero-Party & First-Party Data For Paid Media

Article

Most Brands Ignore Their Best Influencers