The Fractional CMO Show
The Fractional CMO Way explores the evolving world of marketing leadership through the lens of fractional Chief Marketing Officers. Hosted by the experts at RiseOpp, this podcast dives into strategies, success stories, and practical insights that help growing companies scale effectively without the full-time executive overhead. Whether you're a startup founder, a marketing leader, or a business owner looking for high-impact marketing guidance, this show will equip you with the tools and mindset to thrive.
The Fractional CMO Show
Marketing Attribution Tools: Why Most Data Misleads You
Use Left/Right to seek, Home/End to jump to start or end. Hold shift to jump forward or backward.
Why Marketing Data Often Fails explores how marketing attribution tools help businesses understand what actually drives conversions and revenue.
In this podcast, we break down different attribution models, from basic single-touch approaches to advanced multi-touch and data-driven frameworks, and how they impact decision-making.
Whether you are a marketer, founder, or growth leader, you will learn how to choose the right tools, improve data accuracy, and build a clearer picture of your customer journey.
👉 Read the full guide:
The Ultimate Guide to Marketing Attribution Tools
Imagine you're solving uh a multi-million dollar mystery.
SPEAKER_00Aaron Powell Oh, high stakes, I like it.
SPEAKER_02Right. So the crime here is a customer finally deciding to buy your product. But looking at the evidence, you have like a dozen different suspects who could be responsible for making it happen.
SPEAKER_00Aaron Powell You've got the Instagram ad they saw three months ago, right?
SPEAKER_02Exactly. Or you know, a blog post they read, a webinar they attended, an email campaign they clicked, maybe a final sales call.
SPEAKER_00Aaron Powell And the ultimate question is who really did it?
SPEAKER_02Aaron Powell Who actually gets the credit for that revenue? Because today we are answering basically the single most agonizing question in modern business, which is uh which of your marketing efforts are actually driving revenue?
SPEAKER_00Aaron Powell It is the million-dollar question, literally.
SPEAKER_02Aaron Powell To figure this out, we're pulling from this incredibly comprehensive industry text. It's called The Marketer's Guide to Revenue Attribution and Growth Models, and it reviews the exact mechanics of how companies map out your digital footprint.
SPEAKER_00Aaron Powell It really pulls back the curtain on the whole process.
SPEAKER_02Aaron Powell Yeah. And the mission for today's deep dive is to completely shortcut your path to understanding this system. So whether you are, you know, prepping to defend marketing budget or you're just insanely curious about how the targeted ads you see are tracked back to your wallet, we are going to break down the machinery. Trevor Burrus, Jr.
SPEAKER_00Because it is a highly complex machine these days.
SPEAKER_02Okay, let's unpack this. Because for a long time, the way businesses solved this mystery was essentially just grabbing whoever was standing closest to the cash register and handing them a medal.
SPEAKER_00Aaron Powell Which is frankly a terrible way to measure human behavior. I mean, if we look at the data driving this shift like that, think with Google Research highlighted in the guide: 76% of marketers are either currently using sophisticated attribution tools or they plan to within 12 months.
SPEAKER_0276%? That's a massive shift.
SPEAKER_00Aaron Powell The urgency is absolutely there because the customer journey is no longer a straight line, right? It's not just seeing an ad and immediately buying a product. It is a completely tangled web of device switching, passive scrolling, and just, you know, fragmented attention.
SPEAKER_02The journey is a total mess, but executives still demand a clean spreadsheet at the end of the month.
SPEAKER_00Always.
SPEAKER_02And I know the guy breaks down uh five core reasons why getting this measurement right is strategically critical. And it's not just about making marketing look good, right?
SPEAKER_00Far from it. I mean, the first reason is simply about capital allocation, improving investment decisions.
SPEAKER_02Aaron Powell Meaning where to put the money.
SPEAKER_00Exactly. If a business has an extra$10,000 to spend next month, do they pour it into Google search ads or do they sponsor a podcast? Without a mechanism to trace past revenue back to those specific channels, they are literally just gambling.
SPEAKER_02Which is terrifying for a CFO.
SPEAKER_00Right. And the second reason flows right from that clarifying the real nonlinear journey. When you actually map it out with data, you realize people take wildly unpredictable paths to purchase.
SPEAKER_02Which naturally leads to the third piece, right? Connecting those early top-of-funnel interactions to downstream revenue.
SPEAKER_00Yes. Because it is so easy to measure the final click. But it is incredibly hard to prove that, say, a broad brand awareness YouTube video somebody watched back in January is the actual reason they clicked purchase in July.
SPEAKER_02Oh, that six-month time gap has to create so much internal friction.
SPEAKER_00Intense friction. Which actually brings us to the fourth reason: aligning marketing and sales. This is a massive pain point in business to business or B2B environments.
SPEAKER_02Oh, the classic turf war. Marketing teams generate leads, sales teams close them.
SPEAKER_00Exactly. If sales teams think the leads are garbage and marketing thinks sales is just dropping the ball, an attribution system acts as an objective referee. It shows exactly where the handoff worked or where it failed.
SPEAKER_02And the fifth reason.
SPEAKER_00Finally, the fifth reason is strategic accountability for executives. It forces the entire C-suite to speak the same mathematical language.
SPEAKER_02Wait, I have to jump in on that executive accountability piece because if you ask literally any CEO, they already know the buyer's journey isn't a straight line. I mean, they experience it in their own lives as consumers.
SPEAKER_00Of course they do.
SPEAKER_02So why do they still hyperfocus on just that checkout moment when they're reviewing the numbers? It's like we're watching a basketball game, but we're only paying the player who makes the final shot and totally ignoring the three people who pass the ball to get them into position.
SPEAKER_00That is a great analogy. If we connect this to the bigger picture, that hyperfocus happens because companies constantly confuse reporting features with a measurement framework.
SPEAKER_02What do you mean by that?
SPEAKER_00Well, the guy gives a very stern warning about this. You don't just, you know, slip a switch in your analytics software and suddenly have attribution. A true framework requires a massive underlying infrastructure.
SPEAKER_02Like data collection protocols.
SPEAKER_00Exactly. Data collection, identity resolution, so you actually know who is who across devices, statistical modeling, and then crucially strategic interpretation. Most companies skip all that and are stuck looking at what the guide calls activity metrics.
SPEAKER_02Ah, like website traffic or how many likes a post got on LinkedIn. So using the basketball thing, those metrics tell you people are running around on the court sweating, but they don't prove anyone actually scored a point.
SPEAKER_00Precisely. Without a systemic framework to connect activity to revenue, your marketing just remains a collection of disconnected tactics. You have a lot of noise, but absolutely no proof of influence.
SPEAKER_02So let's talk about how companies actually attempt to prove that influence. How do they divide up the credit for those passes? Because the guide goes deep into the specific rules, these attribution models that dictate who gets paid.
SPEAKER_00They do. And the oldest method is still the single touch model.
SPEAKER_02Aaron Powell Which sounds a bit outdated.
SPEAKER_00They are blunt instruments, honestly. You have first touch attribution, which gives 100% of the credit to the very first interaction a customer had with your brand.
SPEAKER_02100% of the first touch.
SPEAKER_00Yeah. The logic is that without that initial introduction, nothing else would have happened. It is highly effective if your only goal is measuring pure brand awareness. But the fatal flaw is that it completely ignores the months of nurturing, education, and follow-up that actually convince the person to buy.
SPEAKER_02So if I click a banner ad by accident in March, I get aggressively courted by a sales rep for six months, I finally buy in September, and the banner ad gets all the glory.
SPEAKER_00Under a first touch model, yes.
SPEAKER_02It makes no sense. And the reverse of that is last touch attribution, right? Where the final trigger gets 100% of the credit.
SPEAKER_00Aaron Powell, which is the default for a lot of legacy systems out there. Last touch is great if you are strictly optimizing the bottom of your funnel. Like maybe you are testing out different colors for the buy now button.
SPEAKER_02Right, highly transactional.
SPEAKER_00But it ignores all the trust building that took place prior to that moment. Both first and last touch are dangerous because they condition companies to overinvest in either the very top of the funnel or the very bottom.
SPEAKER_02While totally starving the middle where the actual persuasion happens.
SPEAKER_00Exactly. You lose the messy middle.
SPEAKER_02Well, if single touch is fundamentally broken, it seems obvious that we need to divide the credit up. And the simplest multi-touch model the text mentions is the linear model, which honestly sounds incredibly fair.
SPEAKER_00On the surface, sure.
SPEAKER_02Yeah, like you have five touch points, you just divide the credit equally, 20% across the board, everyone gets a participation trophy.
SPEAKER_00The math is easy, but the logic is deeply flawed. A linear model assumes all interactions carry the exact same psychological weight.
SPEAKER_01Oh, I see.
SPEAKER_00Right. So a fleeting two-second ad impression while someone is scrolling on their phone gets the exact same revenue credit as a 30-minute intensive product demonstration with a sales engineer.
SPEAKER_02Oh, wow. Yeah, that's absurd.
SPEAKER_00It completely flattens the reality of how human beings make complex decisions.
SPEAKER_02Okay, if a 30-minute demo obviously matters more and usually happens closer to the actual purchase, do companies just weight the credit by recency?
SPEAKER_00They do, and that is called a time decay model. Interactions that happen closer to the point of purchase get a much larger share of the credit.
SPEAKER_02Making the assumption that the later touches are more important.
SPEAKER_00The underlying psychological assumption here is that as buyers get closer to pulling out their credit card, their intent sharpens. They move from casual browsing to active, serious evaluation. So those recent interactions just naturally carry more critical weight.
SPEAKER_02But doesn't that just I mean, isn't that essentially recreating the last touch problem? It feels like it would severely undervalue the early awareness efforts that put the company on the buyer's radar in the first place.
SPEAKER_00It absolutely does, which is why the industry tried to engineer a compromise with what are called position-based models.
SPEAKER_02Right. The guide talks about the U-shaped model.
SPEAKER_00The U-shaped model is probably the most famous of these. It assigns 40% of the credit to the first touch, 40% to the last touch, and then evenly spreads the remaining 20% across everything that happened in the middle.
SPEAKER_02Giving the biggest rewards to the introduction and the close.
SPEAKER_00The strategic argument there is that introducing the brand to a stranger and finally getting them to hand over money are the two hardest hurdles to clear.
SPEAKER_02Okay, I'm looking at this 40, 40, 20 split, and then there's the W-shaped model, which the guide notes is heavily used in B2B. W shaped adds a third 40% milestone right in the middle for league creation or opportunity creation.
SPEAKER_00Yes, the B2B favorite.
SPEAKER_02But I have to push back here. Isn't W-shaped just inventing arbitrary milestones to make B2B sales teams happy and justify their paychecks? Like, does any human being on Earth actually make a purchasing decision in a strict 40, 40, 20 ratio?
SPEAKER_00You're hitting on the core tension of rules-based attribution right there. These models rely entirely on fixed, predefined assumptions.
SPEAKER_01They're just guessing.
SPEAKER_00They're incredibly practical for aligning operations and paying out commissions, but they were blind to reality. Think about it. What if, in your specific journey, a random middle interaction like reading an incredibly detailed white paper was the true deciding factor that won you over?
SPEAKER_02Which happens all the time.
SPEAKER_00Because the U shape or W shape enforces a rigid mathematical rule, it caps the credit that white paper can receive. It overlooks the actual driver of the conversion simply because it didn't fit the pre-assigned percentages.
SPEAKER_02Okay, here's where it gets really interesting. Because if all these fixed rules are just arbitrary guesses at human behavior, why are we guessing at all? Why not just let the math figure it out based on what actually happened?
SPEAKER_00And that brings us to the data-driven revolution.
SPEAKER_02Yes. Letting the data take the wheel.
SPEAKER_00Data-driven attribution is the attempt to completely remove human bias from the equation. It ditches the fixed rules entirely. Instead, it deploys statistical algorithms in machine learning to estimate the true contribution of each touch point.
SPEAKER_02So it adapts.
SPEAKER_00It calculates the credit dynamically based on historical behavioral patterns.
SPEAKER_02The concept sounds great, but how does an algorithm actually know a specific blog post was more influential than, say, a specific ad?
SPEAKER_00By analyzing the failures alongside the successes.
SPEAKER_02Oh, interesting.
SPEAKER_00A data-driven model doesn't just study the paths of people who eventually bought your product, it heavily analyzes the complex patterns of people who didn't buy.
SPEAKER_02The drop-offs.
SPEAKER_00Right. By comparing thousands of journeys that led to closed revenue against thousands of journeys that led nowhere, the machine learning model isolates the true variables.
SPEAKER_02Can you give an example of how that works in practice?
SPEAKER_00Aaron Powell Sure. If a certain webinar consistently appears in the timelines of people who buy, but is notably absent in the timelines of people who bounce, the AI dynamically assigns more weight to that webinar. It calculates the statistical probability of conversion with and without that specific touch point.
SPEAKER_02Aaron Powell Wait, if this AI needs to compare thousands of winning and losing journeys just to find a stable pattern, this can't possibly work for a startup selling 50 products a month.
SPEAKER_00It absolutely doesn't.
SPEAKER_02Oh, really?
SPEAKER_00The volume requirement is massive. The guide explicitly warns that if an organization has fewer than a thousand conversions a month, these data-driven models struggle to find statistical significance.
SPEAKER_02Aaron Powell So the data is just too noisy.
SPEAKER_00Exactly. These systems really only begin to shine when you are feeding them tens of thousands of conversions.
SPEAKER_02Aaron Powell And there's another major challenge with handing this over to AI, right? The black box problem.
SPEAKER_00Aaron Powell Oh, the dreaded black box.
SPEAKER_02Yeah. Because the algorithm distributes the credit, but the marketing team can't actually see the internal logic of how it arrived at that distribution.
SPEAKER_00Aaron Powell It just spits out an answer.
SPEAKER_02Trevor Burrus Right. If my budget is on the line and I have to explain to the CEO why I spent$100,000 on LinkedIn ads last quarter, saying, well, the black box algorithm told me to. Sounds like a terrible defense. How do marketers actually trust a system they can't explain?
SPEAKER_00Aaron Powell The reality is that sophisticated teams don't blindly trust the black box.
SPEAKER_02Okay, good.
SPEAKER_00They use the data-driven model as a highly educated, directional guide, but they never treat it as absolute gospel. They validate the algorithm's recommendations through separate real-world experiments. Makes sense. But you know, to even get to the point where you have a black box to argue over, you need the physical software to track the users in the first place.
SPEAKER_02Let's talk about that tech stack, because these theoretical models have to live inside actual software. And choosing the right tool is rarely about featureless. It's about matching the software to the reality of how your customers buy.
SPEAKER_00Which a lot of companies get wrong.
SPEAKER_02Let's use a metaphor here. If Google Analytics is like a security camera mounted above the front door of your store, HubSpot is more like a private investigator who follows the customer around for six months.
SPEAKER_00That is a highly accurate way to look at the market. Google Analytics 4, or GA4, is the default digital first tool. And interestingly, data-driven attribution is now its default setting out of the box.
SPEAKER_02Which is a big shift for Google.
SPEAKER_00Huge. And as your security camera, it is brilliant at tracking digital movement, especially since it integrates seamlessly with Google Ads. But its major flaw is that it loses the scent incredibly fast.
SPEAKER_01How so?
SPEAKER_00Well, if a customer logs out of their Google account or switches to a different browser, or heaven forbid, picks up the phone to call your sales team, GA4 often just goes blind.
SPEAKER_02Which is why B2B companies need the private investigator. They use tools like HubSpot because in B2B, a single sale might take six months and involve five different decision makers.
SPEAKER_00And GA4 just can't track that long-term complexity.
SPEAKER_02Right. HubSpot bridges the gap because it manages the marketing tracking alongside the customer relationship management or CRM data. It physically connects the ads someone clicked in January to the contract they finally signed in July.
SPEAKER_00Exactly. And then you have the enterprise tier, like Adobe Analytics and its attribution IQ.
SPEAKER_01The heavy hitters.
SPEAKER_00If Google is a security camera, Adobe is a global surveillance network. It is built to handle unimaginably massive data sets. It integrates offline point of sale data, and it allows data scientists to write entirely custom algorithms.
SPEAKER_02Wow. But I'm guessing that's not cheap.
SPEAKER_00Not at all. The catch is that it requires a dedicated, highly paid team just to keep the thing running.
SPEAKER_02Now the guide also highlights a few highly specialized tools. There's ruler analytics, which is engineered specifically for lead gen businesses where the entire journey happens online, but the final sale happens offline over a phone call.
SPEAKER_00Right. It bridges that web cookie directly to the CRM outcome.
SPEAKER_02And then there's Wicked Reports, which caters exclusively to e-commerce.
SPEAKER_00Wicked Reports tackles a very specific business model flaw. In e-commerce, the first purchase a customer makes often barely covers the cost of the advertising it took to acquire them.
SPEAKER_02You're just breaking even on day one.
SPEAKER_00Exactly. The actual profit comes from retention. So Wicked Reports shifts the attribution focus away from the initial transaction and focuses almost entirely on tracking customer lifetime value or LTV.
SPEAKER_02So it attributes marketing spend to repeat purchases over years, not just days.
SPEAKER_00Exactly. It's a completely different measurement philosophy.
SPEAKER_02So you buy the perfect tool for your business model, you plug it in, the problem should be solved. Yet the guide cites a staggering statistic. 70% of marketers still struggle to achieve their goals using attribution data.
SPEAKER_0070%. It's wild.
SPEAKER_02If we have the AI and we have the tailored software, why is this still failing?
SPEAKER_00Because the software is only as good as the environment it operates in. And the text outlines four brutal logistical and human roadblocks.
SPEAKER_02Okay, let's hear them.
SPEAKER_00The first is data fragmentation. Marketing data naturally lives in isolated silos. Facebook hoards its data, your website CRM hoards its data, your billing software has its own completely separate database.
SPEAKER_02They don't want to talk to each other.
SPEAKER_00No, they don't. And if those systems aren't perfectly integrated, the attribution model is essentially analyzing a puzzle with half the pieces missing.
SPEAKER_02The second roadblock is cross-device behavior, right?
SPEAKER_00Yes.
SPEAKER_02So you see an ad on your phone during your morning commute, but you don't click it. Later that night, you open your laptop, search for the brand, and buy the product.
SPEAKER_00A very standard human behavior.
SPEAKER_02But without a reliable way to resolve that identity, the software records that as two completely different human beings. One person saw an ad and vanished. Another person magically appeared from a search engine and bought.
SPEAKER_00And tying those identities together leads directly to the third challenge, which is privacy restrictions. This is a massive headwind right now.
SPEAKER_02Oh, the Apple updates.
SPEAKER_00We are talking about Apple's app tracking transparency. That prompt on your iPhone that suddenly started asking if you wanted to allow apps to track your activity.
SPEAKER_02And most people click no.
SPEAKER_00Overnight, marketers were blinded. Combine that with the ongoing deprecation of third-party cookies across web browsers, and the mechanism companies use to follow you from site to site is just evaporating.
SPEAKER_02If the underlying tracking technology is breaking, how are companies adapting to this? I mean, the guide must offer some solutions to these roadblocks.
SPEAKER_00The primary solution is an aggressive pivot toward first-party data. Companies realize they can no longer rely on Facebook or Google to track you on their behalf. They need to own the relationship directly.
SPEAKER_02Which explains so much of what we see online now.
SPEAKER_00It is the exact mechanism behind why every website you visit now immediately begs you to create an account, subscribe to a newsletter, or join a loyalty program.
SPEAKER_02Just to read an article.
SPEAKER_00Yes, because the moment you log in, they have a persistent first-party identifier, usually your email address, that works across all your devices, completely bypassing Apple or Google's cookie restrictions.
SPEAKER_02It also mentions the need for tracking governance, specifically standardizing UTM parameters, which is basically ensuring that every single link your marketing team puts out on the internet has a standardized name tag attached to it.
SPEAKER_00Governance is incredibly boring, but so critical.
SPEAKER_02Because if the social media team labels a link FB-ad, and the email team labels a link Facebook-promo, the database treats them as totally different channels. It's garbage data in, garbage data out.
SPEAKER_00The governance piece is vital because it feeds into building independent measurement layers. Companies are realizing they cannot let the advertising platforms grade their own homework.
SPEAKER_02What do you mean by grade their own homework?
SPEAKER_00Well, if you look at Facebook's reporting dashboard, it will claim its ads drove 100% of your sales. Look at Google's dashboard, it claims the exact same thing.
SPEAKER_01Right, everyone wants the credit.
SPEAKER_00An independent measurement layer is an overarching system designed to deduplicate those claims and find the actual truth. But establishing that truth triggers the fourth and perhaps most destructive challenge, organizational alignment.
SPEAKER_02You mean office politics? Exactly. Office politics. Attribution exposes incredibly uncomfortable truths.
SPEAKER_00Oh, I can imagine.
SPEAKER_02Imagine you are the director of social media. Under the old reporting system, your channel looked incredibly profitable. But under a new multi-touch data-driven model, the math reveals your ads aren't actually driving new sales, they are just taking credit for demand that was already created by the content team.
SPEAKER_00Yikes. And people's bonuses are tied to this.
SPEAKER_02Budgets, bonuses, promotions. If executive leadership doesn't step in and enforce a unified framework, implementing attribution doesn't bring clarity. It just sparks a civil war between departments.
SPEAKER_00So what does this all mean for you, the person listening to this right now? The text notes that implementing a reliable tracking system takes three to nine months of grueling work.
SPEAKER_02A massive undertaking.
SPEAKER_00Three to nine months just to configure the tracking. I have to ask, are companies just losing the plot here? Is there a point where an organization spends more time and money measuring the marketing than actually doing the marketing? The guide readily admits that organizations overcomplicate this constantly. They build sprawling, hypercomplex dashboards that no one in the company actually understands.
SPEAKER_02Which means no one trusts the data anyway.
SPEAKER_00Exactly. The solution for mature teams isn't to abandon measurement, it's to realize that tracking individual clicks has a ceiling. That is why they layer on advanced macro tactics like marketing mix modeling or MMM.
SPEAKER_01How does MMM bypass the broken tracking cookies and privacy updates?
SPEAKER_00Aaron Ross Powell By ignoring the individual user entirely. Marketing mix modeling is a top-down statistical analysis of aggregated data over long time periods. It looks purely at macro correlations.
SPEAKER_02So it's looking at the forest, not the trees.
SPEAKER_00Right. It asks questions like when we spent an extra$50,000 on television and billboard campaigns in the Pacific Northwest last quarter, do we see a statistically significant spike in overall revenue in that region compared to historical baselines?
SPEAKER_02It doesn't care who specifically bought what.
SPEAKER_00Not at all. It just measures the broader impact of spend on revenue.
SPEAKER_02Which pairs perfectly with the second advanced tactic they mention incrementality testing. I love this concept because it cuts through all the algorithmic noise. It is essentially the ultimate reality check.
SPEAKER_00Attribution models, even the most advanced AI, only estimate influence. They do not prove absolute causation.
SPEAKER_02Right. Incrementality testing is like dealing with a guy who stands right outside the front door of your store handing out$5 coupons to people who are already walking in.
SPEAKER_00Taking credit for the foot traffic.
SPEAKER_02Exactly. At the end of the day, that guy claims he drove a thousand sales because everyone used his coupon, but the coupon didn't cause the sale, it just intercepted it.
SPEAKER_00Which happens in digital marketing all the time.
SPEAKER_02Incrementality testing is when you fire that guy for a month in one specific city to see if sales actually drop. You pause your ads in a controlled geographic region. If your revenue stays exactly the same, those ads weren't driving incremental value. They were just taking credit for people who were going to buy anyway.
SPEAKER_00It is the most humbling test a marketing team can run. It forces the organization to confront the difference between correlation and causation.
SPEAKER_02So as we wrap up this deep dive into the hidden machinery of tracking, what is the ultimate synthesis of all these sources?
SPEAKER_00The core takeaway from the text is that attribution is not about achieving mathematical perfection. Chasing 100% accuracy across millions of data points is a trap.
SPEAKER_01A very expensive trap.
SPEAKER_00Attribution is about clarity. It is an evolving system of continuous learning. You don't just plug the software in and walk away. The ultimate goal is simply to build a framework that equips you to make slightly better, more informed business decisions tomorrow than you made today.
SPEAKER_02I think that's incredibly clarifying. It's an operational compass, not a crystal ball. But I want to leave you, our listener, with a final thought to chew on as you navigate the digital world today. Let's hear it. We've spent this entire deep dive talking about how companies use massive amounts of data and AI to perfectly reconstruct the path that led you to buy a product. They are analyzing your past. But think about where this technology is heading. As these machine learning models consume exponentially more behavioral data, they are moving from analyzing your past to actively predicting your future. At what point does marketing stop being about persuading you to make a choice and start being about engineering your behavior before you even consciously realize you want something?
SPEAKER_00The line between anticipating a need and creating one is getting thinner every day.
SPEAKER_02It really is. Thank you for joining us on this deep dive into the invisible system shaping our choices. We'll catch you next time.