Working With a Creative Agency: What Actually Happens
Save for later
By Nam, Head of Strategy
Published: 24 May 2024 | Last Updated: 14 Sep 2026
There are two ways to work with an agency, and most friction comes from buying one and expecting the other.
Scoped execution. You know what you want. You have a brief, possibly a brand system, sometimes finished designs. You need it built well and on time.
Strategy-led. The brief is open. You have a business problem rather than a specification, and part of the work is deciding what to build.
Both are real engagements. We take both. What causes trouble is a client who commissioned execution and expected strategic input, or one who commissioned strategy and gets frustrated when the brief changes in week three.
This is what each involves, so you can tell which you’re buying.
Scoped execution: what it looks like
You arrive with something defined. A documented brand system and a need for a website. A design file from another studio. A prototype built in Framer or Webflow that now needs to work properly.
What we do: build to your specification, flag risks, deliver on the agreed date.
What we don’t do: relitigate decisions you’ve already made. If your brand guidelines say the button is green, the button is green.
Where we will speak up: when a specification creates a technical problem. A hero video that will fail Core Web Vitals, a layout that breaks with real content lengths, a URL change with no redirect map. We’ll tell you, explain the cost, and build it your way if you still want it.
Typical shape: four to ten weeks, one decision-maker, two revision rounds, a fixed scope document.
Best for: businesses with settled positioning, brand systems already documented, or a specific deliverable with clear boundaries.
Strategy-led: what it looks like
You arrive with a problem rather than a specification. Enquiries have dropped. A competitor repositioned. The business changed and the materials didn’t. You’re launching and don’t yet know how to describe yourself.
What we do: interview across your team, map the competitive set as your buyers see it, test whether the stated problem survives contact with what customers actually say, then design against what we find.
What you should expect: the brief will probably change. Roughly half the time, the problem a client arrives with isn’t the operative one. A company asking for a more premium-looking site often has a pricing-position problem. One asking for a homepage redesign often has a navigation problem three levels down.
Typical shape: eight to sixteen weeks depending on scope, with discovery front-loaded and design starting later than clients expect.
Best for: repositioning, launching, entering a new market, or any situation where two people in your business would describe what you do differently.
How to tell which you need
| Situation | Engagement |
|---|---|
| Brand documented, need a website built | Scoped |
| Design files from another studio, need them built | Scoped |
| One deliverable with clear boundaries | Scoped |
| Prototype exists, needs production build | Scoped |
| Enquiries dropping and nobody knows why | Strategy-led |
| Launching something new | Strategy-led |
| Team describes the business differently | Strategy-led |
| Entering a new market or sector | Strategy-led |
| Brand works but looks dated | Either, depending on whether positioning changed |
If you’re unsure, say so at the first conversation. Scoping a strategy project as execution is the most common way projects go wrong, and it’s visible in week one if anyone asks.
What the first four weeks look like
Week one. Kickoff, access to any systems involved, and agreement on who decides. One name. Every additional approver adds coordination time on top of review time, and a homepage one person signs off in a day can take a committee a fortnight.
Weeks two to three. Discovery on strategy-led work: interviews, competitor mapping, content inventory. On scoped work: technical review, content audit, wireframes.
Week four. First substantive presentation. On strategy projects this is positioning rather than visuals, which surprises clients expecting designs.
Through all of it: one weekly call, one named contact on each side, written summaries after decisions.
Who does what
Clear division prevents most disputes.
We own: creative direction, technical decisions, project sequencing, and telling you when something won’t work.
You own: business decisions, final approval, content and imagery unless commissioned separately, access credentials, and consolidated feedback within agreed turnarounds.
Shared: the definition of success, agreed at the start and measured afterwards.
The content line is the one that causes most overrun. Copy, photography and product data sit directly on the critical path, and content delays are the primary cause in roughly 60% of projects that run late. If you have nothing written, say so at week one rather than week six.
What good feedback looks like
This affects outcomes more than most clients realise.
Consolidate it. One response per round, internally reconciled. Four people commenting across a fortnight is worse than one slower reply that already resolved the disagreements.
Comment on fit, not preference. “I don’t like green” isn’t usable. “Green reads as clinical in our category” is.
Separate the levels. At wireframe stage, comment on structure. At design stage, comment on expression. Aesthetic feedback on a wireframe sends the conversation sideways.
Say when something’s wrong even if you can’t say why. That’s useful information. Diagnosing it is our job.
Warning signs, both directions
Worth naming honestly.
Signs the agency is wrong for you: the brief was accepted without questions; you can’t get a straight answer on who’s doing the work; revisions arrive without explanation of what changed; nobody has mentioned content deadlines.
Signs the client relationship is drifting: approvals routinely late; new stakeholders appearing after decisions were made; scope growing without a conversation about timeline; feedback contradicting previous feedback with no acknowledgment.
Most projects that go badly showed one of these in month one and nobody raised it.
What happens at the end
Handover. Editable source files, exports, documentation, and CMS training where relevant. You own the code, the domain, and the hosting account. Confirm that in writing at proposal stage with any agency.
The first thirty days. Real traffic surfaces what testing didn’t. Monitoring for crawl errors, tracking verification, and fixes.
After that. Either a defined maintenance arrangement or a clean exit. Both are fine. What isn’t fine is nobody owning the site three months in.
Nine in ten of our clients come back, and the reason they give isn’t the design. It’s that the second project is easier, because the system exists and nobody has to re-explain the business.
How we’re set up
Eighteen people in Business Bay, Dubai. Development in-house under technical lead rather than subcontracted, covering custom builds, WordPress, ecommerce and web applications. Strategy under our own strategiest. UI/UX sits inside the brand practice rather than downstream of it, so the people setting the type scale are the people designing the screens it governs.
Since 2000, around 350 projects, for clients including Emaar, Burj Khalifa, VARA, Americana, Al Ghurair, Ajmal Perfumes, Canon, Domino’s, Nissan Middle East and Banque Misr.
Work runs across the UAE, Saudi Arabia, Qatar, Kuwait, Bahrain, Oman, the UK, the United States and India. Remote delivery is normal. The portfolio and industry pages show the range.
Frequently asked questions
A vendor executes a defined specification. A partner helps decide what the specification should be. Both are legitimate engagements with different timelines, costs, and expectations. Problems arise when a client commissions one and expects the other, which is visible in the first conversation if anyone asks.
If your positioning is documented, agreed internally, and still accurate, you need execution. If two people in your business would describe what you do differently, or if you can’t state in one sentence why someone would choose you over the obvious alternative, you need strategy first.
One decision-maker, access to your team for interviews on strategy work, any customer research or win-loss data you hold, consolidated feedback within agreed turnarounds, and content. Budget seventy-five to a hundred and fifty hours across your team for a mid-sized project.
More than most clients expect. Seventy-five to a hundred and fifty hours across three to five people for a mid-sized engagement, covering kickoff, weekly calls, three to five feedback rounds, user acceptance testing and training. It needs to be in someone’s calendar.
Two to three is standard for most deliverables, with additional rounds costed separately. What matters more than the number is how feedback is gathered: consolidated responses use rounds efficiently, fragmented ones burn through them without resolving anything.
Say so, and say what feels wrong even if you can’t articulate why. On strategy-led work, first presentations are often positioning rather than visuals, and a strong negative reaction at that stage is useful rather than a problem. Diagnosing the cause is the agency’s job.
Yes. That’s a scoped execution engagement and it’s common. We’ll flag anything that creates a technical problem, explain the cost, and build it your way if you still want it after that conversation.
You should own final deliverables, editable source files, domain registration and hosting account. Get it in writing at proposal stage. Agencies retaining any of these create lock-in that only becomes visible when you want to leave.
The brief accepted without questions, no straight answer on who is doing the work, revisions arriving without explanation of what changed, and no mention of content deadlines. Most projects that fail showed at least one of these in the first month.
Content that isn’t ready. It’s the primary cause in roughly 60% of projects that overrun. Copy, photography and product data all sit on the critical path, and no amount of development capacity compensates for a page with no text in it.
Yes, as a range. Scope is elastic and the same problem can be solved at several price points. Withholding budget produces proposals that miss in both directions and wastes a round of conversation for everyone.
Raise it immediately rather than at the next review. Scope changes are normal and manageable when they’re discussed; they cause problems when they arrive as feedback disguised as a revision. Around 78% of projects experience some scope creep, so the question is whether it’s handled explicitly.
Not sure which kind of engagement you need? Talk to us and describe the problem rather than the solution. We’ll tell you which it is.
Looking for
Branding Services
?
Call us at +971 50 937 2493
or email us at info@tequila.ae
About the Author
Nam is the research-and-strategy mind behind Tequila's most complex brand work. She gets under the skin of a category — its audiences, competitors, and tensions — and turns what she finds into positioning that actually moves the needle. Designers and clients both lean on her for the same reason: clarity. She makes the hard problems solvable.
Related Articles?
Branding & Web Design Enquiries
working with brands across the UAE and globally.