Luxury advertising budget allocation fails more often than luxury creative does. The brands we see struggling rarely have a spending problem in absolute terms - they have a distribution problem. The money is split across too many channels to reach minimum viable weight in any of them, paced against a fiscal calendar rather than a purchase cycle, and then reallocated every six weeks on the strength of last-click reports that cannot possibly resolve a 90-day consideration window.
This piece lays out the allocation decisions in the order they should be made: how much to spend, how to split it between brand and performance, how to weight channels, how much to hold back for testing, and - the discipline that separates good programs from expensive ones - what evidence should be required before money moves.
How much should a luxury brand spend? Percent-of-revenue ranges
Percent of revenue is a blunt instrument, but it is the right starting constraint because it forces the conversation to begin with unit economics rather than with channel enthusiasm. The ranges below reflect total working media as a share of revenue for brands actively pursuing growth, not maintenance.
| Vertical | Working media as % of revenue | Notes |
|---|---|---|
| Premium and luxury DTC | 12% – 22% | Highest ratio; shortest cycle, most competitive auction dynamics |
| Luxury hospitality (owned property) | 5% – 9% | Excludes OTA commission, which functions as variable distribution cost |
| Luxury automotive (dealer / group) | 1.5% – 3% | Low ratio on high revenue base; co-op funds materially change the math |
| Private aviation (charter, fractional, cards) | 4% – 8% | High LTV justifies upper end; small addressable universe caps the top |
| Luxury real estate (developer / new development) | 2% – 4% of sellout | Front-loaded against release phases rather than annualized |
| Wealth management / RIA | 3% – 6% of revenue | Compliance overhead consumes a real share of total budget |
| Private clubs and membership | 6% – 11% of dues revenue | Concentrated around initiation and renewal windows |
| Luxury goods (watches, jewelry, fashion) | 8% – 15% | Includes significant brand-building obligation |
Two adjustments matter more than the vertical itself. First, customer lifetime value relative to acquisition cost. A brand with a $40,000 LTV and a 4:1 target ratio can support a $10,000 CAC, which changes the whole conversation about what a $60 CPM means. Second, share of voice versus share of market. The most durable finding in advertising effectiveness research is that brands holding excess share of voice - spending a larger share of category advertising than their share of market - tend to gain share over subsequent years, while brands in deficit lose it. If you are a challenger in a category with two entrenched incumbents, the arithmetic requires you to overspend your share position or accept slow decline.
The brand-to-performance split, adjusted for consideration cycle
The widely cited 60/40 brand-to-activation split comes from large-sample effectiveness research across mostly mass-market categories. It is a reasonable anchor and a poor final answer for high-consideration luxury, because the length of the consideration window changes what "activation" can even do.
Our working guidance, based on how long the purchase decision actually takes:
- Sales cycle under 14 days (premium DTC, gifting, some hospitality): 50% brand / 50% performance. Short cycles let response media do real work.
- Sales cycle 30 to 90 days (luxury goods, private club membership, boutique hospitality groups): 60% / 40%. The canonical split holds well here.
- Sales cycle 90 to 180 days (private aviation, luxury automotive, high-end residential): 65% to 70% brand / 30% to 35% performance. Response media has too few in-market prospects at any moment to absorb more.
- Sales cycle beyond 180 days (wealth management, new development real estate, ultra-luxury): 70% to 75% brand. The performance layer becomes a harvesting mechanism, not a growth engine.
The failure mode is universal and predictable: brands with long cycles report that search and retargeting produce the best measured return, so they shift budget there, and for two or three quarters the reported return improves while total volume flattens. What has actually happened is that the harvesting layer is capturing demand created by the brand layer that has just been defunded. This is the single most common way a luxury media program declines while every dashboard says it is improving. It is also why we insist on incrementality testing rather than platform-reported conversions before any structural reallocation.
Channel weights inside the brand and performance layers
Once the split is set, the allocation question becomes how to distribute weight without falling below minimum viable budget in any channel. Spreading $600,000 across nine channels produces nine underweight campaigns and no learnings.
| Channel | Typical share of working media | Minimum viable annual budget | Primary role |
|---|---|---|---|
| Premium CTV (PMP / programmatic guaranteed) | 30% – 45% | $250K | Reach and preference against affluent households |
| Paid search (brand + non-brand) | 10% – 18% | $60K | Demand capture; brand terms are a defensive cost |
| Premium native / contextual display | 8% – 14% | $75K | Sustained presence, article-level context |
| Streaming audio | 6% – 12% | $80K | Efficient frequency, market-level weight |
| Podcasts (host-read or premium spot) | 5% – 12% | $100K | Credibility and narrative length |
| Paid social (organic-adjacent, creative-led) | 6% – 12% | $50K | Creative testing velocity, retargeting |
| DOOH (airports, clubs, luxury retail districts) | 4% – 10% | $120K | Context and proximity; poor as a solo channel |
| YouTube Select | 4% – 9% | $75K | Video reach extension at lower CPM than CTV |
| Test reserve | 8% – 12% | - | See below |
Minimum viable budget is the number most plans violate. Below roughly $250,000 annually, a premium CTV program cannot sustain enough deduplicated household reach to move a brand metric or generate a readable incrementality result. Below about $120,000, DOOH buys reduce to a handful of screens with no measurable frequency. If the total budget cannot support a channel at viable weight, the correct decision is not to buy it thinly - it is not to buy it.
Pacing: always-on, bursts, and the calendar that actually matters
Luxury brands routinely pace media against a fiscal year and then wonder why performance is uneven. The pacing question has three inputs: the purchase cycle, the auction cycle, and the category's demand seasonality.
- Always-on baseline (55% – 70% of budget). High-consideration categories need continuous presence because a small fraction of the universe enters the market each month. A dark month is not a saving; it is a month of demand handed to whoever stayed on.
- Seasonal and event bursts (20% – 30%). Concentrated weight around genuine demand peaks - Q4 gifting, spring travel booking, model year changeover, membership drives, development release phases.
- Opportunistic reserve (8% – 15%). Held for competitive events, unexpected inventory, or market dislocations. Most plans allocate this to zero and then cannibalize the baseline when something arises.
Auction cost seasonality deserves explicit planning. Q4 CPMs in premium video commonly run 25% to 45% above Q1–Q3 averages as retail and mass-market advertisers enter the market. For a brand whose actual demand is not concentrated in Q4, buying flat weight across the year means paying a premium for the least valuable impressions. Shifting 10% to 15% of annual weight out of December and into January and February frequently buys 20% to 30% more delivered reach for the same money.
The test reserve nobody wants to fund
Every luxury media budget should carry an explicit, ring-fenced testing allocation of 8% to 12% of working media. Not "we test within the plan" - a separate line that cannot be raided when a quarter looks soft.
The reserve funds three things:
- Incrementality and holdout tests. Geo holdouts require deliberately withholding spend in matched markets, which means accepting known short-term volume loss to buy causal knowledge. Without a dedicated budget line, this test never survives a planning meeting.
- New channel and inventory pilots. Running a channel at genuine minimum viable weight for two quarters to produce a real read, rather than a $30,000 experiment that proves nothing.
- Creative variance testing. In premium CTV specifically, creative differences between the best and worst-performing assets in a rotation commonly produce a 30% to 60% spread in brand lift outcomes - larger than most targeting refinements deliver.
Brands that fund this reserve consistently develop a compounding advantage, because they are the only ones in their category who know which of their channels actually cause revenue.
When to reallocate: evidence thresholds
The discipline that most improves budget performance is deciding, before the year starts, what evidence will be required to move money. Our standing rules:
| Decision | Minimum evidence required | Minimum observation window |
|---|---|---|
| Shift budget between creative assets | In-flight delivery and completion metrics | 2 – 3 weeks |
| Shift budget between publishers or deals in one channel | Deduplicated reach and cost per qualified reach | 4 – 6 weeks |
| Shift budget between channels | Geo holdout or matched-market incrementality result | 8 – 12 weeks minimum |
| Change the brand / performance split | Incrementality result plus 12+ months of MMM history | 2 quarters |
| Exit a channel entirely | Two consecutive incrementality reads below threshold | 2 quarters |
The row that matters most is the third. Moving budget between channels on the basis of platform-reported conversions is the mechanism by which brand budgets quietly migrate into retargeting. A channel-level reallocation is a causal claim, and it requires causal evidence - which for most luxury brands means a geo-based test, not an attribution model.
Five allocation mistakes that cost the most
- Too many channels, none at viable weight. The most common symptom of a committee-built plan. Consolidating from nine channels to five at proper weight routinely improves measured outcomes with no budget increase.
- Pacing to the fiscal calendar instead of the demand and auction calendars. Paying Q4 premiums for impressions that convert in March.
- Letting the harvesting layer eat the demand-creation layer. Search and retargeting will always report the best return and will always be the wrong place to put marginal budget in a long-cycle category.
- Zero test reserve. A brand with no incrementality history is allocating on opinion, and will keep doing so indefinitely.
- Reallocating faster than the sales cycle. A 120-day consideration window cannot be evaluated in a 30-day review. Judging it that way systematically defunds everything upper-funnel.
Building the allocation
Start from unit economics and set total working media as a share of revenue, sanity-checked against share of voice in the category. Set the brand-to-performance split from the actual measured length of the sales cycle, not from a benchmark borrowed from a shorter-cycle category. Distribute channel weight so that every channel in the plan clears minimum viable budget, and remove the ones that cannot. Pace against demand and auction seasonality rather than the fiscal year. Ring-fence 8% to 12% for testing. And write down, in advance, the evidence standard required to move money - because the pressure to reallocate on weak evidence arrives every single quarter.
If you are rebuilding a luxury media budget and want an allocation grounded in incrementality evidence rather than platform-reported return, apply to work with Stillwater Media. We accept a limited number of engagements each quarter.
