At a glance
Global agency market
Exhibit 1Creative is the weak spot at the big agencies
Organic growth in creative/advertising, Q3 2026 (approximate)
Publicis reports 'low single digit'; that bar is approximate. Omnicom no longer publishes growth by discipline.
The category on a page
Exhibit 2Where creative agencies sit
Profit impact against supply risk, marketing subcategories
Exhibit 4India: digital takes all the growth
2026 forecast growth by medium
Typical savings from an agency fee review
On agency fees and production
Value levers
In order of value.
Ask agencies to show where AI is used and pass on the time saved.
Price by asset or project, not by FTE. AI makes hours a poor measure of value.
Use specialist or in-house production studios for volume content.
With three global groups, run competitive pitches; agencies want share gains.
Make sure content, models and prompts trained on your brand belong to you.
How to buy it
KPIs to track
For the monthly review
Where most buyers are, and what good looks like
| Most buyers today | Best practice | ||
|---|---|---|---|
| Fees | FTE-based retainers | → | Deliverable or outcome-based |
| Production | Inside lead agency | → | Separate production hub or in-house |
| AI | Unknown use | → | Disclosed, savings shared |
| IP | Agency owns tools | → | Client owns assets and models |
| Reviews | Every 5+ years | → | Annual performance review, pitch every 3 years |
What AI changes
More in the AI disruption tab.
From the category playbook
Marketing & creative agencies
| Positioning | Media buying - Leverage · Creative & strategy - Strategic · Production - Leverage |
| Recommended route | Consolidate media into a lead agency (audited, transparent); retain creative relationship; competitive production |
| Spec approach | Scope-of-work & output-based fees (not open retainers); media transparency & rebate clauses; benchmark fees |
| Top value levers | 1) Fee/retainer benchmarking 2) Media transparency & efficiency 3) Roster consolidation & production competition |
| Key risk | Media inflation · fee/rebate opacity · creative switching cost |
| Headline KPIs | Fee vs benchmark · media price vs index · working:non-working ratio · agency scorecard · effectiveness |
Exhibit 5Kraljic positioning (category playbook)
Sub-categories by profit impact and supply risk
Exhibit 6Where the money goes
Typical cost split
Exhibit 7Where the programmatic $ goes - ANA Q2 2026
% of spend
Savings potential (category playbook)
Typical range on addressable spend
Verified benchmarks - Global
- Global ad spend +11.9% to $1.34tn in 2026, moderating to +8.4% in 2027; digital media price inflation running ~9-13% depending on channelT2WARC / WFA · Oct 2026
- Only 45.1% of programmatic spend reaches quality impressions (ANA Q2'26) - a record, but fees still take 27.2%; the leakage dwarfs the rateT2ANA - Programmatic Transparency · Q2 2026
- India media inflation +9.6% - highest of the major markets (vs +4.6% global, +3.9% US)T2WFA - Media inflation outlook · 2026
Supplier landscape - Global
Top global players and the local or regional players that matter here. Illustrative, from public sources; not an endorsement.
Global players
- Omnicom (incl. IPG)World's largest holdco (IPG merger, Nov 2025)
- Publicis GroupeData/commerce-led (Epsilon)
- WPPOgilvy, VML, media (EssenceMediacom)
- Dentsu / HavasOther major networks
Value levers (category playbook)
Marketing spend is mostly media - so the money is in media transparency and the working-vs-non-working ratio, not the creative retainer everyone fights over.
Demand full media transparency (no undisclosed rebates or principal-media arbitrage), audit the media agency, and separate media buying from creative; opacity is where the margin hides.
Measure and shift spend from fees/production (non-working) into actual media (working); a poor ratio means you're paying people more than audiences.
Move creative agencies off fat retainers to project or output-based fees with scope control; retainers drift into paying for idle capacity.
Consolidate the agency roster and decouple production from creative so you're not paying agency mark-up on production you could source direct.
Decide what to in-house (always-on content, performance media) vs keep with agencies (brand, big ideas); in-housing commodity work is a structural saving.
Make AI do the work
Generative AI is collapsing production cost - so renegotiate: production and always-on content that used to justify agency hours now cost a fraction. Claw that back in the fee model and require agencies to disclose where they use AI, so you're not paying human rates for machine output. The brand and strategy work still earns its fee; the production line no longer does.
Reactive vs proactive (category playbook)
| Reactive | Proactive | ||
|---|---|---|---|
| Pricing | Fixed annual retainer | → | Output/SOW with tight scope control |
| Roster | Many agencies, overlap | → | Rationalised roster + lead agency |
| Transparency | Trust the agency's numbers | → | Fee & utilisation transparency; decoupled production |
| Data | Agency owns data & trading | → | In-house the data / trading desk |
| Measurement | Deliverables counted | → | Fee tied to business outcomes |
How AI is changing the category
- Gen-AI reshapes creative and production - faster, cheaper asset generation undercuts the hourly-fee base.
- AI media buying, optimisation and programmatic shift value toward data and algorithms.
- AI fee-benchmarking and media-transparency tools audit agency cost, mark-ups and rebates.
- Synthetic audiences and AI research compress testing and insight cycles.
What to do about it
- Renegotiate scopes and fees as Gen-AI cuts production hours - capture the productivity gain, don't leave it with the agency.
- Demand transparency on AI media tools, data ownership and rebate flows.
- Use AI fee-benchmarking to move to output-based pricing, not FTE/retainer.
- Insert AI-usage, IP ownership and data-governance clauses into the master services agreement.
Each number carries source, as-of date and a credibility tier (T1 official · T2 index/survey · T3 analyst · T4 estimate). Figures are WFA media-price inflation (per market), WARC ad-spend growth and ANA programmatic transparency. Agency fees/rebates are negotiated, not published - benchmark via an independent auditor. India's media inflation is the highest of the major markets, and under half of programmatic spend reaches quality impressions.
Full written analysis
Written live for Creative agencies and your region, grounded on the figures above. Public data only.
Written live for this category and region when the page opens in a browser.
Negotiation playbook
Negotiation playbook
Creative agencies are under pressure: revenue is falling at WPP and Omnicom. Use that to re-base fees and lock in AI savings.
Pick the right commercial model
| Fixed fee per deliverableUsual best | Pay for outputs. Captures AI productivity automatically. |
| Scope-based retainer | Predictable; requires a clear scope and regular true-up. |
| FTE-based retainer | Pays for hours; AI gains stay with the agency. |
| Performance-linked fee | Part of the fee tied to results. |
Questions to ask suppliers
- AI savings share
How much of our work now uses AI tools, and how will that productivity show up in lower rates?
Good answer: productivity gains passed through as lower rates. - Rate card cut
Which roles on the rate card are most affected by AI, and what reduction will you apply to them?
Good answer: for production roles most affected by AI. - IP ownership
Will we own all assets, prompts and any fine-tuned models created for us?
Good answer: all assets, prompts and fine-tuned models. - Disclosure
Where and how will you use AI on our account, and will you disclose it on every deliverable?
Good answer: where and how AI is used. - Exit
If we move agencies, what transition support and asset handover will you commit to, at what cost?
Good answer: transition support and asset handover.
Your leverage
- Pitch competition among three groups and independents
- Volume of content
- In-house alternatives
- Contract term
Watch out for
- Agencies charging for proprietary AI tools on top of fees
- Hours-based billing for AI-produced work
- IP clauses giving agencies rights to models
Media inflation - 2026 (by market)
| Component | Change / forecast | Source |
|---|---|---|
| Media inflation - Global | +4.6% | T2WFA 2026 · 2026 |
| Media inflation - US | +3.9% | T2WFA 2026 · 2026 |
| Media inflation - India | +9.6% | T2WFA 2026 · 2026 |
| Programmatic $ reaching quality | 45.1% | T2ANA Q2'26 · Q2 2026 |
Pricing models (category playbook)
Agency pricing is drifting from retainers to outputs - pick the model that ties spend to value, not headcount.
| Output / deliverable (SOW)Usual best | Pay per asset or campaign, not FTEs - kills retainer bloat and makes spend comparable. |
| Monthly retainer | Predictable, but rewards hours over outcomes; audit utilisation and scope. |
| Commission on media (% of spend) | Legacy model that misaligns incentives toward more spend. |
| Value / performance-based | Fee linked to business outcomes - powerful where attribution is clean. |
Benchmarks to bring to the table - Global
- Global ad spend +11.9% to $1.34tn in 2026, moderating to +8.4% in 2027; digital media price inflation running ~9-13% depending on channelT2WARC / WFA · Oct 2026
- Only 45.1% of programmatic spend reaches quality impressions (ANA Q2'26) - a record, but fees still take 27.2%; the leakage dwarfs the rateT2ANA - Programmatic Transparency · Q2 2026
- India media inflation +9.6% - highest of the major markets (vs +4.6% global, +3.9% US)T2WFA - Media inflation outlook · 2026
AI disruption
Tailwinds for buyers
- Content variations and adaptations in minutes.
- In-house AI studios replace parts of agency production.
- Easier comparison of agency efficiency.
- Fewer agency hours needed for routine work.
- AI testing shows which creative works before launch.
Headwinds for buyers
- Agencies bundle AI platforms as new paid services.
- Brand safety and IP risk from AI content.
- Copyright rules for AI content are still unclear.
- Agencies keep productivity gains if fees are hours-based.
- Fewer agencies with strong AI capability.
How to capture it
Build AI productivity into the fee model before renewal; measure cost per asset.
Who is changing the market
- Brands make content variations in-house with AI, reducing agency production work
- Licensed training data reduces IP risk
- Teams create on-brand design without an agency
- Templates and AI tools replace many small agency jobs
- AI trained on your brand produces copy and images at scale
- Moves content creation from agencies to marketing teams
Category playbook view
Tailwinds
- Generative production drops always-on content to a fraction of the cost - renegotiate the fee for work AI now does.
- AI media optimisation tightens targeting in the ~60% of spend that is media.
- In-housing commodity content becomes viable, removing agency mark-up on routine work.
Headwinds
- Agencies rebrand AI-assisted production as premium creative while their own cost falls - the fee model hides it.
- AI media buying deepens the agency/ad-tech black box unless you demand log-level transparency.
- Brand-safety risk rises as AI-generated and AI-placed inventory scales.
Capture it
Claw back the production hours AI eliminated, and pay full rate only for the brand and strategy thinking it can't do.
Risk radar
Risk radar
Exposure by dimension, 0-100, Global
Risk heat matrix
Impact against likelihood, category playbook; register numbers follow
- Transparency/rebate
- Brand/reputational
- Data privacy
- Agency over-reliance
Top risk flags
- MediumFee / rebate (AVB) transparency
- MediumBrand-safety & reputational misstep
- MediumData-privacy & ad-tech compliance
- LowAgency over-reliance / IP ownership
Risk register
Hours-based fees don't fall when AI cuts effort.
Unclear rights in AI-generated assets.
WPP is cutting jobs; Omnicom integrating IPG.
AI content can be off-brand or wrong.
Market pulse
What the biggest suppliers are reporting
From their latest results and investor material.
Q3 2026. Net revenue €3.81bn; media high single digit, creative low single digit; 2027-28 target +7-8%.
Q3 2026. Core operations organic revenue; integrated media is 53% of core revenue; integrating IPG.
Q3 2026. WPP Creative -5.1%; restructuring under CEO Cindy Rose.
Q3 2026. EMEA +3.5% despite conflict-hit Middle East; Americas -1.2%; Ipsos.Digital (AI, synthetic respondents) +14%.
Latest developments
Dated, newest first. Checked 16 Nov 2026.
WPP Creative -5.1%; Omnicom advertising down; Publicis creative low single-digit growth.
So what: Creative fees should be under review.
The Drum via Haysmac →Record new business; tech practice to return to growth.
So what: Publicis will compete hard for new clients.
Bizcommunity →Adds to 11,000 since start of 2025.
So what: Protect your account team in the contract.
O'Dwyer's →What this means for your next negotiation
- Creative revenue is falling at WPP and Omnicom: re-base fees now.
- Pay per deliverable so AI savings reach you.
- Own the AI assets and models built on your brand.
Signals to watch next quarter
Creative trends.
Account team changes.
Service risk.
Curated developments (category tool)
WARC →
Digiday / WPP Media →
ANA →
ANA / WFA →
Category note
Benchmarks and forecasts - Global
- Global ad spend +11.9% to $1.34tn in 2026, moderating to +8.4% in 2027; digital media price inflation running ~9-13% depending on channelT2WARC / WFA · Oct 2026
- Only 45.1% of programmatic spend reaches quality impressions (ANA Q2'26) - a record, but fees still take 27.2%; the leakage dwarfs the rateT2ANA - Programmatic Transparency · Q2 2026
- India media inflation +9.6% - highest of the major markets (vs +4.6% global, +3.9% US)T2WFA - Media inflation outlook · 2026
Sources
- Holding company Q3 2026 results (The Drum) · Nov 2026
- Publicis Q3 2026 (exchange4media) · Oct 2026
- Pitch Madison Advertising Report 2026 · Feb 2026
Figures checked 16 Nov 2026. Tags: Official = published by the source; Derived = calculated from published figures; Judgement = The Procurement Lens view.