Article5 Way Too Early Trends + Lessons From 2024
Difficult as it is to believe, we’re rapidly approaching the end of Q2 – Back to school media buys are being finalized, BFCM meetings are appearing on the calendar and initial budgeting for 2025 is in the works. Hanging over all of those is a gigantic question: what’s actually going on in the marketing ecosystem? What have we learned thus far in 2024, and (most crucially) how should we apply it to what we’re thinking about in the short (back to school), medium (BFCM) and longer (2025 – onward) term?
As I’m thinking about the answer to this, I keep coming back to a quote typically ascribed to Vladimir Lenin: “There are decades where nothing happens, and there are weeks where decades happen.” Without a doubt, 2024 has had more than its fair share of those weeks.
Perhaps a better way of thinking about it is constructive interference (and you all thought wave dynamics was useless) – for years, we’ve discussed various changes happening in parts of the digital ecosystem: privacy, zero-click, machine learning/generative AI, automation, decentralization, marketing-finance connectivity, the list goes on. Each one is akin to dropping a boulder into a lake – it makes a splash + a wave outward that causes a bigger-than-normal wave to hit the shore. But what’s made 2024 different is that many boulders have been thrown in the pond in rapid succession, and those waves that emanate from each have reinforced each other. Now, instead of a little bigger wave, there’s a tsunami.
What happens in the wake of that tsunami is anyone’s guess. What I can share is what I’m hearing from founders, executives and investors, and what I’m seeing inside client CRMs and Ad Accounts. And thus far, there are 5 trends that have clearly emerged that will define what the landscape looks like in the coming months and years:
1. Efficiency
While most of the last decade has been defined by high growth and cheap money, 2024 has been different. Interest rates are at a near-quarter-century high (the last time rates were this high was Q4 2000 / Q1 2001), with little indication of a change. It’s difficult to overstate the impacts of this in the advertising ecosystem:
- Consumer Spending is increasingly being driven by the top 25% – 30% of households; any lingering pandemic-related savings in bottom-50% households has been consumed by housing, utilities and food costs. For advertisers – especially discretionary advertisers (most D2C, services, travel/leisure, lower-level luxury) – that means more competition for a smaller pool of people who can actually afford what you’re offering.
- Growth Dollars are more expensive than ever before – the era of sky-high valuations + LTV-based acquisition is over. Brands can no longer incinerate money on marketing under the guise of “learnings” or “finding PMF”, then expect to access more cheap capital later. Result: advertisers are demanding more immediate and more substantial returns from marketing investments than ever before.
- Accountability for performance is increasingly stringent – we’re seeing more brands move on from agencies and low-performing employees at a progressively higher clip. Organizations are increasingly turning to automation and GenAI tools (more on that below) to raise productivity for little-to-no marginal cost. On the advertising side, brands are expecting higher returns on their invested capital and are increasingly willing to slash unproductive spend.
- Access To Capital is drying up for all but the most rock-solid people + organizations (who don’t need it). The impacts of this are being felt everywhere, from tighter inventory levels, to lower ad budgets, to reduce expansion/capex. All of that gives entrenched players an edge, since they don’t have to grow at a rapid pace. Even brands that can access capital are reluctant to do so at current rates, which all compounds into slower growth.
- Profit-Taking is on the rise across-the-board. That means fewer discounts (or trading discounts for more financially-favorable offers, like a free gift with purchase or a value-add), more efficiency with ad spend, slower hiring, SaaS consolidation, aggressive contract re-negotiations, increased emphasis on retention and less reliance on debt. Put another way: companies and investors alike are thinking that a slowdown is coming (if it isn’t here already), and are preparing accordingly.
The bottom line is this: we’re in the tightest financial environment experienced by anyone under 40 in their career. While there are outliers – brands growing 50%, 100% YoY – those are the exception, not the rule. Conditions are tougher than they’ve been in a long, long time. Consumers are feeling it. Brands are feeling it.
2. AI-Everything
The overriding theme of both the Meta Performance Summit and Google Marketing Live was “AI-Everything” is here to stay. The GenAI boom feels like what everyone in tech hoped the “Metaverse” would become: a true, paradigm-shifting technology that enabled the next generation of innovation. Major platforms (Google, Microsoft, Meta, Apple) are expected to dump (cumulatively) a trillion dollars (with a “t”) into AI development, cloud, hardware + talent over the next ~5 years. With that eye-watering level of investment comes a commensurate return expectation.
Big Tech’s response: integrate AI into EVERYTHING.
We’re already seeing major changes rolled out in Google:
- Revamps to tCPA (hit in Q2 2024) and tROAS (expected Q3) bidding strategies
- “AI-powered” campaign configuration + ad creative for Google Search Campaigns
- Generative AI has been integrated into Performance Max campaigns
- Rollout of ad units inside AI Overviews on SERPs
- New “immersive”, social-esque shopping ad units via animated image ads in YouTube shorts
- Improved zero/first-party data integrations with Google Ads via Google Ads Data Manager – this is a massive data play, likely in response to increasing data restrictions (more on that below).
- YouTube Shorts – to – shopping ad integration (coming in H2 2024), which allows brands to connect short-form videos from creators to their products in shopping.
Meta has followed suit, with a host of announcements at the 2024 Performance Summit:
- Enabling brands to optimize based on 3P attribution data (Triple Whale, Northbeam, other MMMs, etc.) – this feels like a massive data play to feed their bidding models.
- AI-powered creative tools for ad creation, including text generation + image optimization
- Full AI-powered creative creation tools
- Revamps to Advantage+ campaign types (especially ASC), with an increased emphasis on convincing advertisers to deploy progressively more budget into ASC campaign types.
Obviously, this has been more successful in some areas (data integration, automated campaign setup, some creative applications, some ASC applications) than others (AI Overviews in Search). And while Google has rolled back some of their AI overviews, they’ve been clear that this is a temporary measure. The AI show will go on.
Reading between the lines, it seems that three things are true:
- Platforms are full-on committed to integrating AI into their ad products more rapidly and more comprehensively than most marketers originally believed.
- We’re still in the infancy of these technologies, and platforms are far more willing to accept short-term pain (egg-on-the-face screenshots + user frustration) in return for what they believe is a long-term gain (a transformational technology that will underpin the next 20+ years of search).
- There’s clearly an effort being made to use LLMs + ML to augment data loss caused by rising privacy regulation + legislation. No-one is coming out and admitting it, but Google’s revamp to the Google Ads Data Manager + roll-out of more than 250 CRM integrations, alongside Meta’s pivot to enabling optimization based on 3P attribution + aMMM data has the hallmarks of a “oh $h!4, we need data” realization.
3. Unpredictability
The last point from above dovetails nicely into what we’ve seen across our ad portfolio during H1 2024: massive fluctuations in performance and increasing unpredictability from ad platforms. While ad platforms have always been somewhat volatile, mitigation via offline conversion import, performance targets (cost caps, bid caps, tCPA, tROAS), portfolio bidding, scripts (Google) and stop-loss rules was largely effective.
Candidly, each of those tools individually (save stop-loss rules + hard budget caps) have been less effective (not ineffective, but less effective) during the first part of this year. Tack of rampant platform instability, multiple significant Meta outages, multiple Google delivery + reporting bugs, less reliable data streams (+ data loss), hawk-eyed advertisers ready to pivot/pull back at the first sign of weakness and a shrinking pool of viable consumers and you have a recipe for unstable platform performance. goe
That’s exactly what we’re seeing.
Meta + Google have been more willing than ever before to blow through cost caps + performance targets, especially if other segments within the same ad set/ad group or campaign are outperforming.
Here’s an example from Google Ads Open Research (Austin Pena – definitely worth a follow):
What you can see in the above graph is that Google is bidding way more than it should (CPC bucket is the x-axis) on the right-side of the graph – which is resulting in blowup CPAs (the blue bar). However, since the spend (red bar) is still small, the impacts of those blowups CPAs aren’t obvious in the account. Net effect: Google is able to (effectively) recoup surplus value created in the left and center of the graph (where the blue bar is below the green target line.
Fortunately, there are solutions to this:
Portfolio Bidding in Google Ads, which has an optional setting for advertisers to set a Max CPC bid limit
Bid Caps in Meta, which are quite effective at limiting blowup costs.
Implementing either solution will have the impact of reducing volume in the campaign/account, but that should come as welcome news: the volume (and corresponding spend) was not economically viable to your business, anyway.
Given the realities from (1) above, this is an obvious way that advertisers (agency + in-house) can reduce inefficient/wasted spend AND improve overall account performance, with minimal trade off.
Platform unpredictability/instability has not been limited to spend performance relative to target, though (even if that is the most obvious). The brands that are doing best are the ones that have implemented four major hedges:
1. Data Passback:
Without fail, brands using CAPI (Meta) or Offline Conversion Import + Enhanced Conversions (Google) are seeing better, more stable performance than those relying exclusively on web-triggered events. Similarly, brands with Server-Side tracking are seeing better performance vs. those with browser-triggered events. Data is the fuel that makes your account perform (or not) – so the better and more incremental (i.e. data the platform otherwise would not get) the data you pass to platforms, the higher the probability that those platforms’ ML systems will be able to surface latent patterns and deliver surplus value to you (the advertiser).
2. Consumer Experience:
The entire job of the ad account is to identify a high-potential prospect, prime them with a relevant message, and deliver them to a place where they can take a relevant next step. In effect, that means that the ad platform can do its job perfectly, but if your site/landing page/sales team/warehouse/ops team aren’t up to the task, the outcome of your marketing spend will not accurately reflect the performance of the ad account. Brands that are performing well have remarkable experiences from the instant someone lands on the site through and following the completion of purchase/service delivery. This ensures the optimization result (purchase, qualified lead, etc.) is not disintermediated from the performance of the account.
3. Creative Quantity Drives Creative Quality:
This one is simple: brands that perform best are the ones that embrace the notion that quantity of creative drives the quality of creative. The days of “creative” agencies meticulously designing the “perfect” ad are long-since over (despite what “creative” agencies say). Speed, iteration + volume are superpowers: the brands + agencies that win are the ones who have invested in developing a creative iteration process that allows them to produce more creative per unit time:
Bottom line: creative is an asymmetrical power law game. If you want to win, you need to deploy more creative more efficiently. Remember: 95%+ of your creatives will perform within 2 standard deviations of the mean, and tail distributions occur on both sides – so 97.5%+ of your creatives will perform below 2 standard deviations above the mean. That means that if you launch 2 new ads per week, you’d have a 90% chance of having at least 1 outperformer after 45 weeks, or ~10.4 months.
If you launch 25 new ads per week, you’d need about a month to have that same 90% chance.
And if you launch 100 new ads per week, you’d need only a single week to have that same 90% chance.
What’s more? That knowledge compounds – if you’re able to share 10-15 examples of massive outperformers with your creative team, they’ll be much more likely to identify commonalities among them, which should (in theory) raise your hit rate – which translates to a higher outperformance rate.
4. Integration of Marketing Calendar:
The final element of outperforming brands is a consistent ability to (1) tell a cohesive, compelling brand story that resonates with their target audience (this lifts the brand above the commodity level, allowing pricing power + fostering loyalty) AND (2) connect that story in clear, relevant + meaningful ways to things happening in the real world – events, holidays, happenings (like graduation, or back to school), product/service launches, etc.
This integration is what creates pent-up demand that can be channeled into high converting moments – product drops, events, sales, collaborations, etc. This catalyzes the performance of the ad account (more data!), leading to a virtuous cycle of higher performance, more data, more creative hits, and so on and so forth.
5. (Bonus) Diligent Management + Oversight:
The final ingredient here is one that I often see neglected: you need to be on top of your ad accounts. You need to be on top of your site, your email, your promotions. Media buying isn’t fire-and-forget. You need people with eyes on (and hands off!) your account every day (for high-spending) or every few days (for lower-spend) accounts.
The final point on this is that I would expect – just as we saw with iOS14.5 + iOS15 – that performance will improve as G + Meta’s new data hoovers come online and their ML is able to synthesize new datasets to fill the gaps created by privacy measures/legislation. I fully expect performance on both platforms to materially improve by Q4 (when things really matter for eCommerce), but the seas between now + then will be choppy. To handle that, it might be time to think a bit outside of the Meta + G bubble:
4. Diversification
Let me begin by saying that, all things being equal, I’m an unabashed proponent of focusing on 1-3 core platforms. But things right now aren’t’ equal or normal. The level of unpredictability and uncertainty caused by everything else happening in the ecosystem has me reconsidering this in certain cases.
Consolidation of advertisers into Google + Meta, along with changes to SERPs + regulatory impacts of the looming TikTok ban (even the best TikTok advertisers are only seeing ~10% of sales come from TikTok shops) have resulted in opportunities arising on other, previously neglected + out-of-favor platforms:
- Reddit – any time a site goes from ~1.3B organic visits/year to ~6.9B (est 2024) in a year, it’s likely that CPMs will drop + opportunity will arise for savvy advertisers. That’s exactly what’s happening on Reddit: for the brands who are willing to do their homework and invest in the platform, there is massively efficient customer/client acquisition to be had.
- YouTube + YouTube Shorts – I firmly believe YouTube impressions are some of the most under-valued on the internet – not only is YouTube shorts poised to capture progressively more viewership as people move away from TikTok, but it’s also the only platform that no-one really hates: ~90% of internet users visit YouTube regularly. My 55+ year old HVAC guy had YouTube on his tablet with instructions to fix my compressor thing-a-ma-jiggy, while my niece loves to watch educational videos on it, and I think chef reactions are hilarious. Now, add in increasing viewership from NFL Sunday Ticket + YouTube TV, and you can easily see a pathway to high-value, low-cost impressions.
- LinkedIn – I’ve heard a TON of blowback about LinkedIn over the past 6 months, with many advertisers complaining about sky-high CPCs + lower-quality traffic (esp. via the audience network). But, as user data becomes increasingly muddy with regulation, LinkedIn is poised to become more valuable because 90%+ of its targeting is first-party data. We’ve also seen CPCs + CPMs coming down as advertisers leave – which means there are some value plays to be had for the right brand. It’s not a slam dunk – but it’s an increasing opportunity.
- Pinterest – like Kanye, Pinterest is a former darling that has fallen shockingly far in a short time. But, for the right brand + the right audience, it can be a fantastic top-to-mid funnel channel that drives incremental conversions at an affordable cost. I’ve seen the best performance from Pinterest in categories that have four properties:
- Moderate-to-High Consideration / Research – think home renovations, travel/tourism, fashion/clothing, furniture/furnishings
- “DIY”-ness – recipes, “DIY” home improvement projects, outdoor/lawn care, etc. This extends to products, too – anything with a customizable element (stationary, some makeup/skincare, art projects, etc.)
- High Price Point – this goes hand-in-hand with relatively high consideration, but Pinterest users tend to be planners. For brands seeking the impulsive $35 sale, this probably isn’t the platform.
- Aspirational + Creative – Pinterest is a creativity-focused platform overflowing with beautiful visuals. In order to have any consistent level of success, an aspirational brand + compelling, scroll-stopping visuals is a must.
- Twitter (X) – Where to begin here? Twitter advertising has been in a tailspin since Elon’s takeover – but it’s at the point where dropshippers + course-selling gurus are everywhere…and when those people are on a platform, there’s opportunity. We’ve run some Twitter (X) ads, and the results have been surprisingly good for mid-funnel actions (webinar signups, newsletter sign ups, content downloads) – think $2 to $3 per conversion. The audience is incredibly fractured (X has massive communities in real estate, finance, ecommerce, SaaS, adtech, news/media, politics + more), but has made real strides in improving the ad product + the ad units.
Now, this doesn’t mean every advertiser should try every one of these platforms – but if you’re struggling to get Meta + G to work, this is as good a time as any to run an experimental campaign on one (or more) of the above platforms. I wouldn’t expect it to supplant G + Meta as your primary channels, but any one of these could provide ~10% of the volume with 5% of your budget. In an efficiency-first environment, that could be a worthwhile investment (at least in the short-to-mid term).
5. Regulation + Privacy
Regulation is the most significant and most far-reaching trend shaping the digital ecosystem right now – and the implications of it are so wide-ranging that making predictions on the totality of impact are near-impossible.
Perhaps the best way to discuss these is via the “axes” of impact:
1. Legal Proceedings / Antitrust:
There are (at least) 10 active antitrust proceedings against major ad platform worldwide at the moment:
- Google – has multiple antitrust cases, ranging from (alleged) monopoly power in search, to AdX, to the play store (resolved for $700M) in the US, plus a long-running case related to shopping preference in the EU and an ongoing investigation by the UK CMA, among others.
- Meta – has a running list of antitrust cases in the US (Federal Trade Commission vs. Meta Platforms), Australia (Australian Competition + COnsumer Commission vs. Facebook Inc.), Federal Trade Commission + 46 US Attorneys General vs. Facebook Inc. The final one in the above list was authorized to proceed by the US District Court for the District of Columbia, and seeks to order the divestiture of WhatsApp + Instagram from Facebook.
While most of these cases are proceeding in some way, it’s exceedingly unlikely that any – aside from the AdX case – will result in anything other than large fines and hefty attorneys fees.
Nevertheless, these cases have had a chilling effect in the AdTech space, as being purchased by Meta/Google is increasingly non-viable for disruptive, mid-stage companies under the new US FTC + EU regimes. Ironically, this has the effect of further cementing these two platforms (Meta + Google) as dominant players in the digital ad space.
2. Privacy Regulations:
The second major axis of regulation has been consumer privacy + consumer choice. While historically this was the domain of the EU (and why the EU is increasingly viewed by brands I’ve spoken with as a hostile, lower-priority expansion market), some of the EU’s ideas around user privacy have made their way over to California.
CA has two major consumer privacy regulations: (1) the California Consumer Privacy Act and (2) the California Privacy Rights Act (CPRA). The latter (CPRA) modified the CCPA by adding four additional consumer rights:
- Right To Correction
- Right To Limit Sensitive Information
- Right To Opt Out of Automated Decision Making
- Right To Data Portability
From a business perspective, with each of these new rights comes additional compliance requirements and costs, along with new bottom-feeding law firms looking to make a cheap buck.
Obviously CA is the largest + most visible state with privacy rights, but it’s far from the only one. Virginia, Delaware, Texas, Oregon, Montana, New Hampshire, New Jersey, Tennessee, Kentucky, Iowa, Minnesota, Colorado, Utah, Maryland and Connecticut all have signed at least one piece of consumer privacy legislation into law (source).
State privacy compliance is rapidly becoming a (quite lucrative) cottage industry – and as more states introduce privacy bills (7 states have bills in committee), the probability of a larger, national consumer privacy law increases. For now, the best solution for brands is to comply with the strictest requirement (usually CA, but ask your attorney) while everyone figures out what we’re actually trying to accomplish (other than making a lot of lawyers rich and politicians appear useful).
3. Tech-Sponsored Regulation:
Perhaps the most significant but under-discussed series of regulations are platform-driven ones: think iOS14.5, iOS15, iOS17 (no-one made Apple switch to private browsing). Unlike the other two types of regulation, which require years of hearings + negotiations + discovery, platform-driven regulation just happens with an update.
The same thing has hit Google + Meta, with Google’s Consent Mode v2 and Meta’s Consent Compliance both rolling out in March 2024.
Google is also working directly with the UK Competition + Markets Authority on the phase-out of 3P cookies, which will also have a significant impact on the overall advertising ecosystem (though those effects will largely be felt by smaller platforms, not the big ones).
The challenge with tech-sponsored or tech-driven regulation is that it happens relatively quickly. Sometimes it’s done in response to an investigation or case, sometimes it’s done for a competitive advantage (see Apple), sometimes it is done in response to a perceived legal or legislative threat. But, regardless of the reason, the impacts of these changes are felt primarily by advertisers.
The above says nothing of the pending AI antitrust investigations in the US, or the AI regulation bills already passed in the EU.
Long story long: the rate of tech regulation generally, and ad tech regulation specifically, is accelerating around the world and coming to the US. How that plays out and impacts advertisers is anyone’s guess, but this is a question of when + how, not if.
I hope this is helpful as you think about your short and long-term plans! Until then, enjoy the sunny weather (even if the ad platforms are a bit stormy)!
Cheers,
Sam
