Ask a marketing director what their agency roster costs and you will get a number from the procurement system. Ask what their agency structure costs and the conversation gets much less precise.
The gap between those two numbers is coordination overhead, and in most Thai marketing organisations we have looked at, it is the largest line item nobody has ever costed.
#Where the hours actually go
We ran a time audit with a regional FMCG client operating five partners: a brand agency, a digital agency, a production house, a media agency and a separate performance specialist.
Across one quarter, the internal marketing team of six spent:
- ~11 hours per week in coordination meetings that existed only because the partners could not talk to each other directly
- ~6 hours per week re-briefing the same information to a second or third partner
- ~4 hours per week chasing, converting and re-sharing files between partners
- ~3 hours per week reconciling conflicting recommendations into a single position
That is roughly twenty-four hours a week — most of a full-time role — spent on work that produces nothing a customer will ever see. None of it appears on an invoice. All of it is paid for.
#Cycle time is the metric that exposes it
Cost per deliverable is the number procurement tracks, and it is close to useless for comparing structures. The number that actually reveals the difference is brief-to-live cycle time: the calendar days from an approved brief to the first asset in market.
In a fragmented setup, that clock includes:
- Brief to brand agency, wait for strategic response
- Strategic response to digital agency, re-brief, wait for concepts
- Concepts to production house, re-brief, wait for schedule
- Assets to media agency, wait for trafficking
- Performance specialist discovers the assets are the wrong ratios
Each handover is a queue, and queues do not add — they compound. The same work under one roof skips every step in that list, because the person doing step three was in the room for step one.
Across the consolidations we have run, cycle time falls by roughly forty percent. Not because anyone works faster, but because nobody waits.
#Fragmentation costs consistency before it costs money
Five partners will produce five sincere, competent, subtly different interpretations of your brand. Each is defensible in isolation. Together they read as a brand that is not quite sure what it is.
This shows up first in the places nobody reviews closely — marketplace listings, community replies, always-on social, the third variant of a banner. The hero film is always on brand. The two hundred assets around it are where consistency is actually won or lost.
A single production team that has read the same brief and shares the same asset library produces variance you can measure in percentage points. Five teams produce variance you can see from across the room.
#What consolidation does not fix
It would be dishonest to present this as a clean win.
- Per-deliverable rates rarely drop. You are buying speed and coherence, not a discount. If anything, a good integrated team costs slightly more per unit.
- Depth can thin out. A one-stop team that claims world-class expertise in every discipline is overselling. Keep specialists where depth genuinely beats integration.
- Dependency risk is real. One partner failing is a bigger problem than one of five failing. Mitigate structurally, not emotionally: own your ad accounts, own your data, insist on documented handover, and keep one specialist relationship live as a market reference.
#The structure we recommend
A single team owning the core loop — strategy, creative, content production, media, measurement — with two or three specialist relationships maintained at the edges for genuinely niche work.
That gives you one brief, one accountable partner and one cycle time for eighty percent of the volume, while preserving depth and a competitive reference point where it matters.
#How to test it without betting the year
Run one brand or one market as a pilot for two quarters. Track three numbers before and after:
| Metric | How to measure |
|---|---|
| Brief-to-live cycle time | Calendar days, approved brief to first live asset |
| Internal coordination hours | Time-audit the marketing team for two weeks |
| Variants per production day | Count usable assets produced per shoot day |
If all three move in the right direction, you have evidence rather than a theory. If they do not, you have lost one brand's quarter and learned something specific about your own organisation — which is a considerably cheaper lesson than restructuring the whole roster on a hunch.