Your agency has a Slack channel about you.
Not a mean one. Not a "can you believe this client" channel. More like a running thread where your account lead posts the brief you sent on Friday at 4:47pm and the team tries to figure out what you actually need – because the brief is thorough, well-written, and still doesn't answer the one question that matters: who is this for?
I know this because I've been both of you.
The view from both chairs
Before I started Qontour, I spent decades moving between agency roles and product-side positions – Apple, Microsoft, Salesforce, Zscaler, and others. I was the client who sent feedback that amounted to "it doesn't feel right" and then couldn't understand why the agency needed more direction than that. I hired agencies for their expertise and then sat in rooms where the internal team overrode every recommendation. I've been told I was the best client an agency ever had. Looking back, there were definitely times when that wasn't true.
Now I run a Webflow agency that works with cybersecurity and deeptech brands. I sit on the other side of those calls. I interpret the feedback, navigate stakeholder dynamics, and watch partnerships either strengthen or slowly corrode over misunderstandings that both sides could have prevented.
That dual vantage point is why I'm writing this. Not to tell you what you're doing wrong – you're good at your job, you don't need that from me. But to show you what those familiar partnership moments look like from the other chair. Because once you see it, the friction starts making a lot more sense.
The patterns nobody names
There are specific moments in every agency engagement that clients and agencies experience completely differently. Not because either side is doing something wrong – because you're living the same project from fundamentally different vantage points.
Here's what I mean.
The brief that answers everything except the question
You spent real time on it. Product specs, competitive landscape, messaging pillars, brand guidelines. The agency received a document that demonstrates how well you know your product – and still doesn't tell them who they're writing for, what objection they're trying to overcome, what the buyer's decision process looks like, or what success means in six months. The brief is a product document. The agency needs a marketing document. Those are different things, and the gap between them is where the first round of "misaligned creative" comes from.
The review round that multiplies
Ten stakeholders on a copy review. Each one has a different opinion. A few have veto power. Nobody owns the final call. The agency receives feedback that contradicts itself – one person wants it more technical, another wants it more accessible, a third wants to know why it doesn't look like what Palo Alto is doing. The agency doesn't know which voice to prioritize, so they try to address everything. The next round gets the same treatment. By round four, the original strategy has been revised into something nobody loves but everyone can tolerate.
This isn't a "too many cooks" cliché. It's a structural problem. The agency can't solve it because it's not their org chart. But you can – by deciding before the review who owns the final call and making that explicit to everyone in the room.
The feedback that's a feeling
"It doesn't feel right." "Can we make it more dynamic?" "This isn't quite landing." Every agency has received this feedback. Every client has given it. And both sides know it's not actionable – but nobody says that out loud.
Here's what's usually happening: you can see that something's off, but you don't have the vocabulary to name it in design or copy terms. That's not a failure on your part – it's a different skill set. The fix isn't learning design language. It's giving the agency permission to ask you diagnostic questions instead of treating your feedback as a direct instruction. "It doesn't feel right" is a starting point for a conversation, not a revision note.
The competitive example from another universe
You send the agency three competitor sites you admire. All of them belong to companies with 50x your budget, a 40-person marketing team, and a brand that's been refined over a decade. The agency now has to figure out how to deliver that caliber of work within your actual scope – or explain why the comparison doesn't apply without sounding like they're making excuses.
Better move: send them examples and be specific about what you like. "I like how this company explains a complex product on their homepage" is useful. "Make ours look like CrowdStrike's" is a trap.
The technical briefing that never happened
You hired an agency partly because they said they understand cybersecurity. And they do – at a category level. They know the landscape, the buyer psychology, the compliance environment. What they don't know is the specific technical distinction that makes your product different from the three competitors it gets confused with. The difference between SASE and SSE. Why your approach to data classification matters. What a CISO actually evaluates during a proof of concept.
If the agency doesn't get a real technical briefing from someone on your team who can explain the product in depth, they'll write messaging that's technically accurate and strategically generic. Not because they're lazy – because they don't have the raw material to do better.
The AI layer
This is all getting more complicated. AI is reshaping what clients expect from agencies and what agencies promise they can deliver – and those two things are diverging fast.
Your agency says they're "agentic." Your CEO read something about AI-generated websites. Your board wants to know why the rebrand is taking twelve weeks when they saw a demo that built a landing page in fourteen minutes. Meanwhile, your agency is genuinely integrating AI into parts of their workflow, but the productivity gains don't map neatly onto the timeline or budget reductions your team is expecting.
Nobody in the room has the same definition of what AI actually changes about the engagement. That's a relationship problem before it's a technology problem. And it's one that neither side is addressing directly because the client doesn't want to sound uninformed and the agency doesn't want to sound like they're defending their fees.
The fix is the same as every other pattern here: name it. Have the conversation about what AI changes and what it doesn't before the project starts, not after the first invoice arrives and someone on your team asks why it cost that much when "AI should make this faster."
What shifts when you see it
None of these patterns are hard to fix. A brief that includes the audience and the success metric alongside the product specs. A review process with a named decision-maker. Feedback conversations instead of feedback documents. Competitive references with specific callouts instead of aspirational comparisons. A 30-minute technical briefing before the agency writes a word.
Small structural changes. The kind of stuff that doesn't feel strategic but quietly determines whether your agency partnership produces great work or expensive rounds of mediocre revisions.
I didn't learn any of this from a framework or a book. I learned it by being the client who created these problems without knowing it, and then becoming the agency that receives them. Both perspectives changed how I work. I'm guessing at least one of these patterns will feel familiar to you, too.
The good news: the agencies who work with cybersecurity brands are generally sharp, technically curious people who want to do good work for you. Most of the friction isn't about capability on either side. It's about the gap between two experiences of the same project – and that gap is getting easier to close as more people on both sides start talking about it openly.
Queries
How do I write an agency brief that actually works?
Include the audience (who you're trying to reach), the objection you're trying to overcome, what success looks like in measurable terms, and any competitive context that matters. Product specs are necessary but not sufficient – the agency needs marketing context as much as technical depth.
How many people should review creative work?
Fewer than you think. The most effective review structures have a single decision-maker with input from two or three stakeholders whose feedback is advisory, not binding. Make the decision hierarchy explicit before the first review round – the agency needs to know whose voice to prioritize.
How do I give useful feedback when something doesn't feel right?
Start by describing the reaction rather than prescribing the fix. "The homepage feels too generic" is more useful than "make it pop." Then let the agency ask diagnostic questions – they're trained to translate feelings into design and copy decisions. The conversation is the feedback, not the document.
Should I give my agency a technical briefing on our product?
Yes – and not from marketing. Have someone from engineering or product spend 30 minutes walking the agency through what makes your product technically distinct from the competitors it's most often confused with. This is the single highest-ROI investment you can make in the quality of their output.
How should I talk to my agency about AI and timelines?
Directly, before the project starts. Ask what parts of their process involve AI, how it affects the timeline, and where the human work still lives. This prevents the expectation gap that surfaces later as sticker shock or scope pressure. If your agency can't answer these questions clearly, that's worth knowing early.
