I still remember the first time a friend texted me a photo of a drink that looked like it had been mixed by a cartoon. Pink on one side, blue on the other, a little ridiculous, and somehow already sold out at the store near her office. She was not asking me about coffee. She was asking whether I had seen it yet. That tiny question is the whole game. Most marketing still shouts. The campaigns that actually move a register whisper through someone else’s phone, then disappear before you have time to bargain. I have watched plenty of brands chase that feeling and end up with a coupon. The interesting ones refuse the coupon.
Three very different companies are trying to turn that feeling into a business, not a moment. One sells coffee and is rebuilding trust cup by cup. One sells off-price clothes and home goods and wants you to walk into a store you did not plan to visit. One sells credit cards and is spending heavily to win customers who might stay for years. On the surface they have almost nothing in common. Underneath, they are running the same experiment: can attention, timed correctly and measured honestly, become full-price demand?
Why Buzz Only Counts When It Shows Up In The Register
Attention is cheap to talk about and expensive to own. A clip can rack up views and still leave the parking lot empty. I have found that the brands worth studying are the ones that treat a viral spike as a hypothesis, not a victory lap. Did people try the product? Did they come back the following week? Did the visit happen at full price, or did the company have to bribe them through the door?
That last question matters more than most social dashboards admit. Discounting can manufacture a crowd. It can also train a crowd to wait. Product novelty, a short window, and a story people want to retell do something different. They give a reason to show up while the margin stays intact. Analysts who follow consumer companies have started calling this kind of push holistic, because it mixes culture, urgency, and price discipline in the same move. I would put it more plainly. It is marketing that does not apologize for charging what the menu already says.
The harder part is proof. Brand affinity, consideration, and purchase intent can all rise while the weekly traffic chart stays stubborn. Executives love the softer numbers because they move first. Investors should love the slower ones: transactions, visit frequency, rewards engagement, and comparable sales. If those do not follow, the campaign was theater.
Cultural noise is only useful if someone still cares on a Tuesday. The real test is whether people come back.
Consumer research commentary
Perhaps the most interesting aspect of this moment is how split the playbook has become. One company is keeping marketing near two percent of sales and trying to make each dollar feel louder. Another keeps the budget under one percent and leans on shoppers to advertise the store for free. A third just lifted marketing expense by nearly a quarter, into the billions, and is asking shareholders to treat that outlay like a factory. Same word. Three completely different bets.
A Drink That Refused To Stay A Memory
The pink-and-blue frappuccino first appeared years ago, tied to a mythical animal and a very real line out the door. It could have stayed a 2017 footnote. Instead, social chatter brought it back into the conversation earlier this year, and the company did something sharper than a nationwide relaunch on day one. It parked the drink at a major music festival in April, let the cameras do part of the work, then opened a global window for a single weekend in August.
Two million of those drinks moved across those days, according to the company. That is not a rounding error on a slow Tuesday. It is a concentrated burst of trial, mostly at full price, attached to a broader effort to remind people what the stores are supposed to feel like again. The limited window did the psychological work a coupon usually tries to do. Scarcity without a markdown. I like that trade, when it is real scarcity and not a fake countdown timer on a website.
There is a turnaround sitting underneath the whipped cream. The chief executive, a little over two years into the job and already known for reviving another fast-casual chain, has been blunt about where the money goes. The company is putting spend where it thinks it can drive transactions, rebuild affection, and make the brand legible again. On a recent quarterly call he said marketing sits a little above two percent of sales, that returns are not fading, and that the budget should grow as the business grows. That is a restrained sentence. It is also a claim shareholders can check.
The check, so far, is mixed in the way real recoveries usually are. Global comparable sales have been positive for four straight quarters, with margin expansion in the last two. Brand affinity, consideration, and purchase intent hit five-year highs in the latest quarter, and customer connection improved year over year. Shares are up about 12.5 percent this year, and more than 20 percent since the hiring announcement in mid-August 2024, even after slipping off a 52-week high above $110. Marketing did not do that alone. Menu fixes, store operations, and leadership tone all sit in the same pile. Still, the drink weekend is a clean example of the new posture: invent a reason to visit, refuse the discount, measure what happens next.
Employees As The Least Suspicious Influencers
The unconventional piece is the staff. A Green Apron creator program has been paying baristas to make social content on dedicated shifts. A separate short-video pilot is being prepared that would pay one employee and one outside creator to travel between stores and film coffee, products, and the rooms themselves. On paper that sounds like every brand’s creator deck from the last five years. In practice it is a bet on who the audience already trusts.
Younger buyers are not naive about paid posts. They grew up inside them. One analyst put it well when she said that demographic carries a keen suspicion of authenticity, because so much of what reaches them is obviously purchased. A barista who actually pulls the shots has a head start that a studio spot rarely earns. The accent is real. The complaint about a rushed morning is real. The pride in a drink that came out right is real. Paid media can imitate that. It rarely matches it.
I would not romanticize it. Employee content can go sideways, and a program that feels like homework will read like homework. The version that works gives people room to sound like themselves while the product stays recognizable. If the pilot is mostly travel b-roll with a logo in the corner, it will die in the feed. If it feels like a person who works there showing you why a particular store is worth the detour, it might do what a national ad cannot: make one location feel specific.
- Trial still has to convert into a second visit, not a screenshot.
- Rewards engagement is an early tell that the moment stuck.
- Frequency matters more than a single weekend spike.
- Transactions and comparable sales are the receipts.
- Full price is the point. A viral discount is just a cheaper crowd.
The next report card lands next month, with fiscal fourth-quarter results and, more importantly, a look at how management frames the coming year. Guidance will tell you whether the creative energy is a chapter or a habit. I will be watching the traffic line more than the mood line. Love is nice. Repeat purchases pay the rent.
Marketing As An Offensive Weapon, Not A Brochure
Shift the aisle. The off-price retailer that runs T.J. Maxx, Marshalls, and HomeGoods has a chief executive who has called marketing an offensive weapon since he took the top job in 2016. The phrase is a little martial for a store full of candles and sneakers, but the intent is clear. The spend is supposed to recruit new shoppers, pull existing ones back more often, and take share from rivals who still think a weekly circular is a strategy.
Here is the twist that trips people up. The budget is still under one percent of sales. They are not flooding the zone with money. They are changing the mix. Over the last few years the weight has moved hard toward online, paid creators included. An analyst who covers the group said it plainly: it is not that they are spending more, it is that the recipe changed. Digital and social now carry work that used to sit in broader brand ads about recognizable labels for less.
The free layer might be the more powerful one. Shoppers film their finds. A haul on a short-video app does not need to produce an online order. Most of this company’s sales still happen inside four walls. The clip only has to make someone drive over. That is an awkward truth for pure e-commerce stories and a gift for a retailer whose inventory changes by the hour. You cannot screenshot your way into a size that might be gone by Saturday. You have to go.
Management said the brands generated roughly 1.4 billion paid video views across major social platforms in the first half of the year. Completion rates on two of the biggest video platforms ran well above industry benchmarks, which is a polite way of saying people did not swipe away. I treat view counts with suspicion. Completion is harder to fake. If someone watches the end of a clip about a jacket, they were at least briefly interested in the jacket.
From Bargain Message To Treasure Hunt
The words changed too. For a long time the advertising leaned on value: known brands, lower tickets. That still matters. It is no longer the whole sentence. The newer message is the hunt itself, the chance of finding something that will not be there next time. Analysts say that shift has helped the company speak to higher-income shoppers who want both a deal and merchandise that feels a step up. Price and taste, in the same bag.
Behind the ads sits a quieter muscle. The company is better than most at mining data and reacting, according to longtime retail watchers. Which creative actually pulled a visit. Which channel paid for itself. Which one merely looked busy. Money moves toward the first group and away from the third. That sounds obvious. Plenty of marketing teams still cannot do it, because the data is messy and the agency relationship is political. Being willing to kill a pretty campaign is a competitive advantage dressed up as a spreadsheet.
The budget rule is elegant when growth is healthy. Marketing is managed as a percentage of sales, so the absolute dollars rise as the business rises, without an automatic hit to margins. Share gains over recent years have done exactly that. Comparable sales have run ahead of many competitors, the budget has scaled with them, and investors have mostly shrugged because the sales showed up. That is the deal. Spend can grow. It has to earn the growth.
The deal frays if sales stall. Then leadership either cuts the weapon or spends a higher share of revenue to keep traffic moving. Either choice raises a question. Was the earlier success the marketing, or was the marketing riding a product cycle that has cooled? Shares are down roughly 13.5 percent this year, with a rough patch after the latest earnings report. The largest division, which houses the two big apparel banners, was soft. Guidance was conservative. The chief executive blamed execution, not demand: the right goods were not in the right stores. Fixes, he said, were already in place, and the current quarter had started stronger. Results for that quarter are due around mid-November.
I tend to believe execution explanations when the category itself is not falling apart, and off-price has been taking share for a reason. I also want to see the third-quarter traffic before I treat the explanation as closed. Marketing cannot rescue empty racks. It can make a corrected assortment feel urgent again. That is the job over the next few months: keep visits coming without a meaningful jump in the spend rate. If the treasure-hunt story still pulls people in once the product is fixed, the offensive-weapon line holds. If it does not, the views were just views.
| Company posture | Marketing intensity | What the buzz is for | Proof investors should demand |
| Coffee turnaround | A little above 2% of sales | Full-price trial and brand reset | Frequency, transactions, comps |
| Off-price retailer | Under 1% of sales | Store visits from social hauls | Comp sales and share gains |
| Card issuer | Up 23% to $1.7 billion in one quarter | Long-lived high spenders | Originations and lifetime profit |
When Marketing Starts To Look Like Capex
The third story does not live in a cafe or a clothing rack. It lives in a credit decision. In the second quarter, marketing expense at the card company jumped 23 percent to $1.7 billion. Part of that is the Discover deal closed in May 2025. Part of it is higher direct marketing and media on the legacy book, plus richer premium benefits. The line item is wider than a commercial. It includes the bonuses and rewards that actually convince someone to apply.
Those bonuses are the real bill. One bank analyst called the upfront incentive, especially for heavy spenders, the single biggest cost by a wide margin. Another estimated that landing a high-quality super-prime cardholder can run past $1,000. Read that again. A thousand dollars to win one customer, before that customer has revolved a balance or paid a full year of fees. It sounds reckless until you remember what a loyal high spender is worth if they stay. Loan growth may be slower than with riskier borrowers. The relationship, if it holds, can be far more valuable.
That is why one analyst described the firm as a brand and marketing-led company, and why management thinks of the outlay the way an industrial firm thinks of capital expenditure. A factory produces goods for years after the concrete dries. A well-chosen cardholder can produce interchange, fees, and balances for years after the bonus posts. I have always thought this analogy is both fair and dangerous. Fair, because the payoff is delayed. Dangerous, because a factory shows up on the balance sheet and a campaign does not. You cannot tour the ad.
Acquiring a customer today can be the plant you never get to walk through. The return shows up later, in accounts that stay.
Shares are down 19 percent this year, including a difficult September, as rate-sensitive financials absorbed pressure after the central bank lifted its target range by a quarter point to 3.75 to 4 percent. A soft jobs report on Friday cooled expectations for another hike in October, which should take some heat off the group. None of that fixes the marketing question. It only changes the weather around it. The company still has to show that today’s elevated acquisition cost becomes tomorrow’s profitable relationship.
Lounges, Dinners, And The End Of The Simple Ad
Finding those customers no longer means a mailbox and a television spot, though both still matter. The net is wider: digital ads, social, and experiences that are hard to copy in a banner. Airport lounges. College-sports packages with a VIP pass. Athlete meet-and-greets. Concert access. Cardholder dinners with chefs from serious restaurants. The point is differentiation you can feel, not a slightly better cash-back decimal.
An analyst who follows the name said the era of setting up an email and buying television is long gone. I think that undersells how stubborn old channels are. Direct mail still works on certain households, and a well-cut television ad can still plant a name. What has changed is that those tools are no longer sufficient on their own, especially for a premium customer who is choosing among several metal cards that all promise roughly the same airport convenience. The experience has to be specific enough to remember on a Wednesday.
Artificial intelligence sits in the background of this spend, not as a slogan but as a testing machine. Faster campaign trials. Offers that look a little more like the person receiving them. Used well, that should lower waste. Used badly, it will personalize junk. The technology does not absolve the strategy. It only shortens the loop between a bad idea and the data that kills it.
Measurement is the honest problem. There is no line in the statements that isolates the profit from one celebrity dinner or one lounge opening. Outsiders are asked, in effect, to trust the pattern: strong new-account originations, gains among high spenders, relationships that stay profitable after the bonus year. A significant trust-us element is baked in, and anyone who pretends otherwise is selling a cleaner story than the business allows. I would rather hear management talk about cohort profit than about brand love. Love does not have a loss rate. Cohorts do.
Three Budgets, Three Definitions Of Return
Put the three side by side and the word marketing starts to feel too small. At the coffee company it is a cultural spark plug inside a turnaround, capped near two percent of sales, judged by whether people return at full price. At the off-price retailer it is a light, targeted push under one percent of sales, amplified by customers who film their own bags. At the card issuer it is a multi-billion-dollar acquisition engine, closer to building capacity than to buying awareness.
The shared discipline, when it exists, is refusal. Refusal to confuse a view with a visit. Refusal to let a discount do the work a product should do. Refusal to keep funding a channel just because last year’s plan had it in the deck. I have sat through enough brand reviews to know how rare that refusal is. Teams fall in love with the asset. The asset did not fall in love with the customer.
A practical scorecard: Spike in attention Trial at full price Second visit within a month Rewards or account activity Comparable sales or cohort profit Spend still inside the guardrail
Run a campaign through that list and a lot of famous work falls out. The festival drink clears the first two steps almost by design. The open question is step three. The haul video clears attention and, if the racks are right, can clear the visit. It fails if the product miss returns. The card bonus clears acquisition and then waits, sometimes for years, to prove it was not an expensive hello.
What Younger Buyers Actually Punish
There is a generational wrinkle worth sitting with. The suspicion of paid content is not a moral stance so much as a trained reflex. If every other video in a feed is a partnership, the unmarked one gets a longer look. Brands that borrow an employee’s real day are trying to rent that longer look. Brands that stage an employee’s real day are trying to counterfeit it. People can tell. Not always immediately. Often enough.
The off-price haul is interesting because the poster is usually not on payroll. The incentive is social, not contractual: look what I found, look what I paid. That is harder to industrialize and easier to trust. The card dinner is the opposite end. It is obviously paid for. Its job is not stealth. Its job is to be worth talking about after the plates are cleared. Different contracts with the audience. Both can work if the company does not lie about which contract it signed.
I keep coming back to a simple filter. Would a skeptical twenty-six-year-old send this to a friend without adding a joke about ads? If yes, you have something. If the joke is the whole message, you have a campaign that entertained the agency.
Urgency Without Training People To Wait
Limited windows are fashionable, and fashion is how good ideas get ruined. A weekend-only drink works because the product is genuinely absent afterward. A fake timer on a webpage that resets at midnight works until it does not. Customers learn the pattern. Then urgency becomes a genre, like a horror trailer, and people stop flinching.
The off-price version of urgency is structural. The item really might be gone. You cannot reorder the exact rack. That is an advantage no campaign has to invent, which is why the messaging shift toward the hunt feels more durable than a seasonal stunt. The coffee version has to be invented, carefully, so it does not become the only reason anyone visits. A brand that needs a mythical drink every quarter to make its numbers has not fixed the ordinary cup. The ordinary cup is the business. The unicorn is the invitation.
Card benefits have their own exhaustion risk. Lounges get crowded. Meet-and-greets get posted by everyone. A dinner stops being a story once three issuers host one. The companies that stay ahead will rotate the experience faster than the feed gets bored, and they will be honest when a benefit has become table stakes. Table stakes should be funded like maintenance, not celebrated like invention.
How To Read The Next Set Of Numbers
For the coffee name, the fiscal fourth quarter and the year-ahead outlook are the cleanest near-term test. I want comparable sales that do not depend on one weekend, margins that keep the expansion, and a marketing ratio that stays near the current band unless management can show why more spend buys more than noise. Four positive quarters is a start. It is not a finished turnaround. Leadership has done this kind of repair before, at another chain. Repeating it is the assignment, not the résumé.
For the off-price group, mid-November matters because the chief executive already claimed the fix is visible inside the quarter. If the large division firms up without a jump in marketing intensity, the data-led mix is doing its job. If the quarter needs a louder ad budget to look acceptable, the earlier share gains deserve a second look. Product in the right store is not a marketing sentence. It is an operating sentence. Ads cannot substitute for it for long.
For the card issuer, ignore the headline marketing number in isolation. A 23 percent increase is neither proof of ambition nor proof of waste. Ask what arrived with it. New accounts. Quality of those accounts. Early spend. Retention after the bonus period. Premium benefits that show up in usage, not just in a press announcement. Rate moves will shove the stock around. They will not tell you whether the $1,000 customer was a bargain.
- Separate the spike from the baseline before you applaud a campaign.
- Check whether the visit or the account happened at a healthy margin.
- Look one period past the launch, when novelty has faded.
- Compare spend growth with sales growth, not with last year’s deck.
- Treat management anecdotes as hypotheses until the cohort shows up.
The Quiet Risk In A Loud Campaign
Every one of these bets has a failure mode that does not look like failure at first. The coffee brand can win a weekend and lose the month if stores still feel inconsistent. The off-price retailer can flood social feeds and still miss the quarter if buyers mis-locate the goods. The card company can originate a beautiful book of accounts that never earns back the bonus once rewards get competed away. Buzz hides those problems for a while. It does not retire them.
There is also a market-level risk that has nothing to do with creative. A weaker shopper, a rate path that stays tight, a jobs report that stops being a one-off, any of those can make a good campaign look ineffective. That is why I prefer companies that can point to full-price demand and repeat behavior, not just reach. Reach collapses first when households get careful. Habit lasts longer.
In my experience, the investors who get this wrong are the ones who fall in love with the anecdote. The festival. The billion views. The chef’s table. Anecdotes are how the strategy becomes legible. They are not the strategy. The strategy is a repeatable way to create a reason to transact, at a cost the margin can carry, with a measurement loop rude enough to kill what does not work.
What A Good Marketing Dollar Feels Like Now
A good dollar in this environment is specific. It knows which customer it wants and what that customer already ignores. It borrows urgency from the product instead of printing a fake deadline. It shows up in a channel the buyer actually uses, including, awkwardly, the employee’s own phone. It leaves a trace in the data: a trial, a haul that becomes a visit, an application that becomes a second year of spend.
A bad dollar is loud, broad, and proud of itself. It buys a familiar face, a familiar song, and a familiar shrug. It reports impressions. It asks for patience. Sometimes patience is warranted, especially in card economics, where the payoff is built to be late. Patience without a cohort is just hope with a media plan.
I do not think the old channels are dead. I think they have been demoted. Mail still finds certain households. Television still builds a name when the name needs building. What no longer works is pretending those tools are the whole machine. The machine now includes a weekend product drop, a creator who actually works the shift, a shopper filming a cart, a lounge that has to be worth the detour, and a model that tells you which of those earned the next dollar.
Useful test: attention x urgency x full price x repeat = a campaign that belongs in the model. Drop any factor and you mostly have content.Where I Would Place The Emphasis
If I had to rank the clarity of the link between buzz and business, the off-price story is the cleanest on a normal day. The customer already wants the hunt. Social simply points at the door. The budget is small enough that a miss does not rewrite the margin. The risk is operational, which is annoying and also fixable.
The coffee story is the most entertaining and the most dependent on follow-through. Two million drinks is a fact. A durable traffic recovery is a sequence. I like the refusal to discount. I like the employee angle, with reservations. I will like it more if the ordinary week starts to resemble the special weekend.
The card story is the hardest to score and the easiest to overpay for. Treating acquisition as capex is intellectually honest. It also asks for a longer leash than most shareholders enjoy giving. The experiences are memorable. Memorable is not the same as incremental. Until the cohorts speak, the $1.7 billion is a claim.
None of this requires a person to pick a winner in public. It requires a person to stop treating marketing as a vague brand cloud and start treating it as a set of mechanisms with different clocks. One clock runs in days. One runs in store visits. One runs in years. Mix those clocks up and you will praise the wrong quarter.
A Note On Skepticism That Still Leaves Room
Skepticism is the right default. It is not the same as cynicism. Cynicism assumes the drink, the haul, and the dinner are all empty. They are not empty. People really did queue. People really do film their finds. People really do choose a card because the lounge made a delay feel less grim. The work is to ask what those moments are attached to six weeks later.
Companies are getting better data on that question than they had a decade ago, which is why the excuses should be getting thinner. If you can see which ad drove a sale, you can stop funding the one that did not. If you can see that employee content holds attention longer than a polished spot, you can shift the mix without writing a manifesto about authenticity. The tools are less mystical than the keynote suggests. The discipline to use them is the scarce part.
That discipline shows up in small sentences on earnings calls. We are not seeing diminishing returns. We moved the mix online. We think of this like capital spending. Those sentences are worth underlining, then checking against the next print. A management team that describes the mechanism clearly has at least started the job. A management team that then shows the mechanism in the numbers has done it.
The Part That Still Feels Human
Strip away the dashboards and you are left with a person deciding whether to cross a street. The drink photo is a nudge. The haul is a nudge. The lounge photo from a friend’s trip is a nudge. Brand marketing, at its least pompous, is the craft of making that nudge feel like information rather than a trap. I trust the version that lets me leave without a coupon code. I trust it more when the store, the rack, or the account is still worth it after the story fades.
Maybe that is why the old frappuccino landed again. It was silly enough to send and specific enough to buy. The companies copying the shape of that moment, without the product or the window or the follow-up, will get the screenshot and miss the sale. The ones that keep the full price, tell the truth about who is speaking, and then look at the register instead of the applause will turn buzz into something sturdier.
I do not need every campaign to be clever. I need the clever ones to survive contact with a normal week. That is the bar. A mythical drink can clear it once. A treasure hunt can clear it most Saturdays. A card benefit can clear it if the second year still looks good. Everything else is content. Content is fine. It is not a business.
Watch the next prints with that distinction in your pocket. If transactions, visits, and cohorts confirm the story, the marketing was an investment. If they do not, it was a very expensive way to be talked about. Plenty of brands will settle for being talked about. The ones worth owning are pickier than that.
