William & Mary — Paid Media Strategy Review
A Media Plan
Built On Intent.
An audit of the current OHO approach and Sway's intent-based, persona-driven alternative — built on the creative foundation we've already built together.
Current annual spend
$248K
OHO FY26 plan
Proposed budget
$450K
Hypothetical planning figure
Avg Tier 1A applications
~7,600
Students per year from these markets
OHO social spend — demographic
57%
$23,250 of $41K total social budget
OHO budget in paid search
76%
$191,500 of $251K · Google text ads only
OHO paid search — branded
40%
$76K defending interest that already exists
Context
Sway is the AOR on creative and branding for W&M. We've built the messaging framework, imagery, and ad mechanic testing — with full creative development as the natural next step. This plan is how all of that work reaches the right student, in the right channel, at the right moment in their decision process.
Media Audit — Targeting Methodology
OHO's $248K — What's Being Bought
76% of OHO's budget is in Google Search text ads — a format where every school looks identical and W&M cannot differentiate.
Google Search text
$191,500
76% · text ads only
YouTube visual
$12,500
5% · only visual intent channel
Demographic social
$23,250
9% · age + HHI targeting
List-based
$13,500
5% · social + yield
Retargeting
$10,250
4% · Snap + YouTube
Google Search text — $191,500
YouTube visual — $12,500
Demographic social — $23,250
List-based — $13,500
Retargeting — $10,250
Google Search breakdown
Branded search
$76,000
42% of search spend · high % of spend
Tours, visits, student life, admissions. Capturing clicks mostly driven by existing brand awareness. Limited competitive value when the student is already searching W&M specifically.
Non-branded search
$105,500
58% of search spend · competitive
Programs ($85.5K) + general non-brand ($20K). Text ads indistinguishable from every competitor — and broad keyword match risks attracting searchers who never fit W&M's admissions profile.
What $105,500 in non-brand search can't do
Differentiate W&M from the 50 other schools running identical text ads in the same auction. Communicate anything beyond a headline and two lines of copy. Tell a story. Build brand preference. Win the moment — not just show up in it.
The real problem
OHO's plan does nothing for the junior whose consideration set is being built right now — through visual, brand-expressive channels that text search cannot reach. And even when it does intercept the right student, a 90-character text ad is not how W&M's story gets told.
Sway Strategic Framework
The Enrollment Pipeline
Five distinct audiences. Five distinct jobs. Senior awareness (demo-based) shouldn't exist in your plan.
Awareness — intent-based, not yet W&M aware
Consideration — engaged with W&M brand
Intent-based consideration — comparing W&M vs. competitors
Yield — applicants & accepted students, list-based
Rising junior summer — first touch, underserved
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Key dates
Common App
ED1 Deadline
ED1 Decisions
ED2 + RD
ED2 Decisions
RD Decisions
Decision Day
Sway Plan — Intent-Based · Competitive Schools Tier (UVA, Duke, Georgetown, UNC, etc.)
Rising juniors
Juniors — unaware
Juniors — engaged
Seniors
Parents
Key shift
Senior awareness (demo-based) shouldn't exist in your plan. If the junior pipeline works, seniors already know W&M — the paid media job shifts to intent-based consideration for applicants, and bottom-of-funnel conversion for accepted students. Same channels, different message, different moment. The rising junior summer window is the highest-opportunity, lowest-competition moment in the cycle — OHO's monthly spend here drops as low as $250/month across all Tier 1A targets.
Media Audit — Creative Differentiation
Intent Gets You In Front.
The Format Determines If You Win.
The vast majority of OHO's intent-based spend is in channels where creative differentiation is severely limited — with YouTube as the only exception, and at $12.5K it can't build meaningful frequency.
OHO's current channels
Google Search text
Zero ceiling$181,500
One headline. One URL. Two lines of copy. Every school looks identical in 90 characters. No visual. No voice. No brand differentiation possible.
YouTube keyword targeting
Underfunded$12,500
Right format — video allows brand expression. But ~$1K/month across a full year can't build meaningful frequency. Scattered impressions don't accumulate into brand preference.
Snapchat + Instagram demo
Wrong targeting$23,250
Visual format has real differentiation potential. The constraint is the targeting mechanism — demographic age + HHI, rather than intent-based behavioral signals.
Programmatic display
Not in plan$0
No programmatic display spend across the full FY26 plan — no behavioral targeting via Trade Desk, no open web retargeting, and no parent-reach across streaming or display inventory.
Sway's recommended channels
TikTok — intent-based (policy permitting)
High ceilingTo confirm
7.36% engagement rate for higher ed. Full video canvas. Persona-matched creative. Behavioral intent targeting. Subject to W&M institutional policy — flagged for discussion.
YouTube pre-roll — properly funded
High ceiling
Campus feel, student life, program depth — the right format for W&M's story. Needs meaningful budget to build frequency. Particularly powerful for junior awareness and program-specific layers.
Instagram — LiveRamp 3P intent targeting
High ceiling
Sway's LiveRamp integration deploys 3rd-party intent audiences directly into Meta — moving beyond Meta's 1P demographic targeting to reach students based on demonstrated research behavior rather than age and income.
Snapchat — list-based, engaged
High ceiling
Snapchat is the right channel for the consideration stage — admitted students Feb–May. The fix isn't the platform, it's the targeting: list-based against Carnegie/Slate admits rather than broad demographic age targeting.
Programmatic display (Trade Desk)
High ceilingNew to W&M plan
Cross-device retargeting to reach student targets' parents — deploy Carnegie/Slate list audiences via Trade Desk across display inventory that OHO's plan doesn't touch.
Google Search — kept + enhanced with conquest
Foundation preserved
Sway keeps the Search foundation and adds competitor conquest (flagged in the current plan's roadmap, never executed) plus Sway Signals DMA-level budget prioritization — with exact match keyword targeting to minimize waste from unqualified audiences.
The Strategic Gap
The opportunity is bringing intent-based targeting to high-ceiling formats — and through our LiveRamp integration, deploying 3rd-party intent audiences across social channels with precision that Meta's native 1P targeting can't replicate. Same channels, fundamentally different targeting architecture. The tested messaging work lands where it was built to land.
Proprietary Intelligence
Sway Signals: The Data Layer
Real-time search demand intelligence across all 210 US DMAs — built specifically to tell universities where active student interest is forming before it shows up in application data.
What it measures
Live search demand by keyword cluster
Sway Signals tracks search volume for any keyword cluster — competitor school names, program terms, college comparison queries — across every DMA simultaneously. Updated monthly, not quarterly or annually.
12–18 months ahead of application data
When a junior in Raleigh starts searching "Duke vs Georgetown" in October, that intent signal appears in Sway Signals months before any application is filed. It's a leading indicator — not a lagging one.
Competitive share of voice by market
Sway Signals shows W&M's share of active consideration search against competitors — by state and DMA. Where share is high in a historically weak market (low applications), we're winning awareness but losing the conversion game — that's a messaging and funnel problem. Where share is low in a historically strong market, we're not even in the consideration set — that's the highest-priority prospecting target.
How universities use it
Budget concentration — not spray
Instead of setting broad geographic tiers and running uniform prospecting spend across them, Signals identifies the specific DMAs where live demand is high — so budget concentrates where students are actively researching, not where the map assumes they should be.
Competitor-specific creative triggers
A student in Charlotte searching UNC Chapel Hill gets a different W&M message than one in Raleigh searching Duke. Signals tells us which competitor is winning the consideration set in each market — so creative can counter it directly.
Emerging market discovery
Signals surfaces DMAs across all 210 markets where W&M-relevant demand is building before it appears in enrollment data. New feeder markets get identified and tested before competitors even notice them.
Why it matters
Most paid media plans for higher ed are built on lagging signals — application history, enrollment data, last year's geo performance. Sway Signals puts W&M's media budget where students are researching right now — 12 to 18 months before that interest becomes an application.
Media Audit — Geographic Strategy
Where to Concentrate.
Not Where to Spray.
Application history tells you which markets have worked. Sway Signals tells you which of those markets have active demand right now. Together they identify where intent-based prospecting spend should concentrate — and where it will dilute.
The dilution problem
OHO targets New York, Texas, Florida, and North Carolina at the whole-state level. These are enormous, economically diverse geographies. Demo-based prospecting cannot work effectively across an entire state — the signal is too thin and the spend spreads across markets with wildly different demand density.
Sway's approach — signals + history = concentration
Historical application data shows which markets have produced results. Sway Signals shows which DMAs within those markets have live search demand for competitive schools right now. The overlap — proven market, active demand — is where prospecting budget concentrates.
This also surfaces emerging DMAs across all 210 markets where demand is building before it appears in application data — letting Sway act on intent 12–18 months ahead of the historical signal.
Note: List-based marketing (Carnegie/Slate, admitted students) operates differently and can be deployed across any market. The geo concentration argument applies specifically to intent-based prospecting.
W&M share of active consideration search by state (vs. Duke, UNC, UVA, Georgetown)
Where W&M's share is low in a historically strong market, active demand exists but W&M is underrepresented in the consideration set. These are the highest-priority DMAs for concentrated prospecting — not whole-state coverage.
Source: Sway Signals
The play
North Carolina is W&M's 5th largest application market — competitors generate 34× more search volume there than W&M. Statewide targeting spreads budget from Charlotte to Asheville. Sway Signals identifies the specific DMAs where that consideration is live — so prospecting spend concentrates where the demand signal is, not where the state line ends.
Sway 1P Data Framework
List-Based & Engaged Audiences
The difference between a media plan that reaches students and one that reaches the right students — is the data architecture underneath it.
List taxonomy — where audiences come from
Prospecting lists
Carnegie/Slate name purchases and inquiry records — the starting pool. Segmented by engagement level: purchased name only vs. genuine RFI submission. Name-buy-only records stay in awareness; RFI submitters advance to consideration targeting.
Intent-based lists
Students who requested information directly from W&M — visited campus, completed an inquiry form, or responded to outreach. Highest intent signal in the funnel. These audiences get consideration and competitive-comparison messaging, not awareness.
Pixel-based custom audiences
Built from W&M site behavior — retargeting visitors who didn't apply, users who clicked contact or info forms, and program-page engagers. Separate pools by action type so messaging matches where they stopped.
Parent audiences
Separate list logic targeting parents via Trade Desk display and streaming. Decision-makers and influencers require different channels, different creative, and different timing than the student themselves.
Activation architecture & list hygiene
LiveRamp / Trade Desk activation
All lists push into LiveRamp for identity resolution and live deployment across Meta, Snapchat, YouTube, and Trade Desk programmatic. Lists update continuously — not a one-time upload — so the audience in-platform always reflects current CRM state.
Multi-platform activation
The same list hits the student across channels simultaneously — Meta, Snapchat, YouTube, Trade Desk display. Frequency is controlled at the list level, not the platform level, so spend doesn't stack against the same person across channels unintentionally.
Age segmentation & funnel timing
Lists are segmented by age from point of first touch. A record that's 18 months old gets a different message — or gets suppressed — versus one from last week. Recency tiers prevent stale audiences from diluting performance and ensure funnel logic stays accurate over time.
Suppression & exclusions
Enrolled students, converted admits, and junk inquiries (purchased name-only with zero engagement) are excluded from active targeting. Suppression lists protect budget and keep frequency away from audiences who've already made their decision.
* Subject to refinement based on discovery with W&M and tech integration capabilities.
Why this
matters
Most agencies upload a list and forget it. Sway's list architecture is live, segmented, and suppression-aware — which means the audience in-platform gets smarter every week, not stale. That's the difference between a list-based campaign and a list-based system.
Sway Hyper Segmentation
Moneyball for Universities.
Out-precision, not out-spend. Every ad is a function of market, audience, competitive set, and persona — served at the DMA/Region level, built on live intent data, and matched to creative that speaks to that student specifically.
The four-layer targeting stack
Layer 1 — Market
DMA-level targeting based on Sway Signals demand data. Budget concentrates where active consideration is live. Different DMAs get different creative as research tells us which markets respond to which themes.
Layer 2 — Audience & Funnel Stage
Rising junior, junior unaware, junior engaged, senior in consideration, yield-stage admit — each gets a distinct message, channel mix, and cadence. Same student, different moment, different job for the ad to do.
Layer 3 — Competitive Set
Sway Signals identifies which competitor schools a student is actively researching. A student comparing W&M to UVA gets a different message than one comparing to Georgetown or Duke — because the decision calculus is different.
Layer 4 — Persona
The messaging framework Sway built maps to four W&M personas. Creative isn't interchangeable — an Ambitious Architect and a Curious Eclectic respond to fundamentally different proof points, even within the same DMA and the same funnel stage.
A worked example — one student, one ad decision
Market
DC / Baltimore DMA
High W&M application volume · Sway Signals demand active
Audience
Junior — Unaware
In prospecting pool · No prior W&M engagement
Intent signal
Public-Ivy Research
Actively searching UVA, Georgetown, W&M comparisons
Persona
Ambitious Architect
Prestige-driven · Wants elite outcomes without the crowd
The ad that runs
"Mega-universities: You're just a number. Small schools: Limited opportunities. W&M: Personal attention, unlimited potential."
Channel: YouTube pre-roll + Instagram (LiveRamp intent audience) · Message focus: Academics / Outcomes · Persona: Ambitious Architect · CTA: Explore W&M
As ad mechanic testing data comes in, we learn which markets respond to which themes — and the creative matrix updates. The plan gets smarter every flight, not just every year.
* Hyper-segmentation depth (2–4 layers) is subject to platform targeting capabilities and available audience scale at the time of activation.
The edge
Competitors spend more on paid media. W&M doesn't need to match that — it needs to be more precise. Four targeting layers means every dollar is doing a specific job for a specific student in a specific market. That's not a media plan. That's a system.
Strategic Summary
OHO vs. Sway
OHO's timing is defensible. The case for Sway is everything inside the calendar that's not being done.
Topic
OHO approach
Sway approach
Awareness targeting
OHO —Demo-based (age 15–17 + HHI) on Snapchat and Instagram. Hopes the right student is in the bucket.
Sway —Intent-based via Sway Signals. Students demonstrating active college research behavior. Every impression pre-qualified.
Junior strategy
OHO —"Jr/Sr Combined" throughout. Juniors and seniors treated identically. No pipeline logic.
Sway —Three distinct junior audiences: rising juniors (first touch), unaware (full-year awareness), engaged (consideration).
List architecture
OHO —Static list uploads. No segmentation by recency, engagement level, or funnel stage.
Sway —Live lists via LiveRamp — segmented by intent level, age, recency, and suppression. Updated continuously, not one-time uploads.
Hyper-segmentation
OHO —Appears to run one message per channel with no persona, region, or competitive-set logic applied to creative.*
Sway —Up to four-layer targeting: market/region + audience + competitive set + persona. Different creative per combination.
Message architecture
OHO —No visible connection to a tested messaging framework, persona architecture, or funnel-stage logic.
Sway —Tested messaging pillars mapped to personas, funnel stage, and channel. The ad mechanic work informs every impression.
Competitor conquest
OHO —OHO's own roadmap flags it. Never built.
Sway —Built from day one. Messages vary by region — and potentially by DMA — based on Sway Signals demand data.
Market intelligence
OHO —Static "Tier 1A Geos" — undefined, unchanging, application-volume-based allocation.
Sway —Sway Signals across 210 DMAs. Real-time search demand. Budget concentrates where active consideration lives.
* Based on available plan documentation. Actual OHO creative segmentation may vary.
The core argument
A generic higher ed playbook is table stakes. It doesn't win the student. Sway is the only agency that can run a plan where the message, audience, channel, and creative all pull in the same direction — because the hyper-segmentation, live data infrastructure, and creative architecture required to do it right is exactly what most agencies push back on. We're built for it.
Proprietary Intelligence
Sway Signals: The Unfair Advantage
Real-time search demand across all 210 US DMAs — turning geo assumptions into geo intelligence.
What Sway Signals measures
Search demand volume for any keyword cluster — competitor school names, program terms, college comparison queries — across all 210 US DMAs simultaneously. Updated in real time. 12–18 months ahead of application data as a leading indicator of where new W&M prospects are forming.
Market entry example — North Carolina
Raleigh-Durham: high "Duke admissions" search demand → lead with W&M value vs. Duke cost
Charlotte: high "UNC Chapel Hill" demand → lead with W&M personal attention vs. large state school anonymity
One state. Two completely different campaigns. OHO runs the same message everywhere.
Keyword clusters to run in Sway Signals
Competitor conquest
UVA admissionsDuke acceptance rateGeorgetown vs W&MUNC Chapel HillDartmouth admissionsEmory admissionsVanderbilt requirements
Summer junior discovery
Best public universities east coastPublic ivy listTop liberal arts collegesBest value collegesColleges like UVASmall class size universities
Application research
W&M early action deadlineW&M acceptance rateColleges 1300 SATW&M financial aidW&M pre-lawW&M CS ranking
Yield / decision
W&M vs Georgetown redditShould I go to W&MW&M vs UVAW&M dormsW&M scholarshipsLife at William and Mary
The pitch in one sentence
Sway is the only agency with the creative architecture, the proprietary demand intelligence, and the persona framework to build a media plan where every dollar connects to a qualified audience, a tested message, and a channel with a high differentiation ceiling. OHO can buy media. Only Sway can buy W&M's media.