The worst feedback I ever gave was technically correct.
It was 2024. We were about six months into PipelineRoad. A contractor we’d brought on for content had delivered a batch of blog posts for a client. The posts were well-written in isolation but completely missed the client’s voice. They sounded like generic B2B content — the kind you could paste any logo onto without anyone noticing.
I wrote a long message. Detailed. Point-by-point. I explained exactly what was wrong with each post, referenced the brand guide, cited specific examples of where the tone was off. I was thorough and precise and, I thought, helpful.
She quit the next day.
Not because the feedback was wrong. Because it was delivered like an autopsy report. Clinical, exhaustive, and devoid of any acknowledgment that she’d spent twenty hours on work she was proud of. I’d given her a spreadsheet when she needed a conversation.
That failure taught me more about leadership than any book I’ve read.
Why Most Feedback Fails
Most feedback fails for one of three reasons, and they’re all about the giver, not the receiver.
Timing. You give it too late, when the moment has passed and the person has already moved on mentally. Or you give it too early, when emotions are still hot and neither party can think clearly. The worst version is giving feedback months after the fact, during a performance review, when the person has no way to connect the feedback to the actual moment.
Framing. You frame it as a verdict instead of a direction. “This isn’t good enough” is a verdict. “Here’s what would make this land with the client” is a direction. The first one stops the conversation. The second one starts it.
Asymmetry. You have context the other person doesn’t have. You know what the client said in a private call. You know the competitive landscape. You know the strategic intent behind the brief. But you give feedback as if the other person has all this context, and when they don’t meet your expectations, you attribute it to carelessness instead of information gaps.
Every feedback failure I’ve had — and I’ve had many — can be traced to one of these three.
The Framework
Over the past two years, I’ve developed a feedback approach that I use with everyone at PipelineRoad. It’s not original — it borrows from things I’ve read, conversations with other founders, and a lot of trial and error. But it works for us.
Step one: Context before critique.
Before I say anything about the work, I make sure the person understands the context I’m operating from. What did the client say? What’s the strategic priority? What constraints exist that might not be obvious?
This sounds basic. It isn’t. Most of the time when I think someone did a bad job, what actually happened is they did a reasonable job based on the information they had. The gap isn’t in their skill — it’s in my communication of what was needed.
I’d estimate that 40% of the feedback I’m about to give dissolves once I share the context. The person sees the gap immediately, without me having to name it. That’s the ideal outcome — self-correction through better information.
Step two: Anchor on the specific, not the general.
“The tone is off” is useless. “In paragraph three, the phrase ‘unlock your potential’ sounds like a motivational poster — this client’s audience is CFOs who think in spreadsheets, so we need language that sounds like it belongs in a board deck” is useful.
Specific feedback gives people something to work with. General feedback gives them something to worry about. The difference between “this needs work” and “the CTA in email four assumes the reader has already agreed to a meeting, but at this stage they’re still evaluating — we need a softer ask” is the difference between confusion and clarity.
I keep a running list of before/after examples for common feedback patterns. When I’m reviewing work, I reference those examples. It makes the feedback feel less personal and more technical, which is exactly the right framing for creative work.
Step three: Separate the work from the worker.
This is the hardest part and the one I still get wrong sometimes.
Creative people — writers, designers, strategists — invest themselves in their work. When you critique the work, they feel it personally. You can’t eliminate that entirely, but you can manage it.
The language matters more than you think. “This post needs a different angle” is work-focused. “You missed the point” is person-focused. They might mean the same thing to you, but they land very differently.
I’ve trained myself to never use “you” when giving feedback on work. It’s always “this section,” “the headline,” “the email flow.” Never “you wrote,” “you missed,” “you should have.” It’s a small linguistic shift that changes the entire emotional register of the conversation.
Step four: Close the loop.
Feedback isn’t a monologue. The worst thing you can do is deliver a wall of notes and walk away. The person needs to respond — to ask questions, push back, explain their thinking.
Some of my best strategic insights have come from the pushback step. A designer explains why they made a choice I didn’t understand, and I realize the brief was ambiguous. A writer defends a headline and convinces me it’s actually better than what I had in mind. The feedback conversation isn’t just about correction — it’s about calibration. Both people leave with a more refined understanding of what good looks like.
I always end feedback conversations with a version of: “Does this make sense? What am I missing?” Not as a politeness — as a genuine question. Because sometimes I am missing something.
Cultural Complexity
PipelineRoad’s team is spread across multiple countries and cultures. This makes feedback more nuanced than most founders appreciate.
The directness that works with one person can feel aggressive to another, not because of individual sensitivity, but because of deeply ingrained cultural norms about hierarchy, face-saving, and communication style.
I learned this the hard way with a team member who consistently agreed with my feedback in the moment and then didn’t implement it. At first, I was frustrated. Was he ignoring me? Was he careless?
Neither. In his cultural context, directly disagreeing with a manager was deeply uncomfortable. His “yes” didn’t mean agreement — it meant acknowledgment. The actual disagreement was expressed through inaction, which was the only safe way he knew to signal that something wasn’t right.
Once I understood that, I changed my approach with him. Instead of presenting feedback as conclusions, I started presenting it as questions. “What would happen if we tried this angle instead?” “How do you think the client would respond to this framing?” It gave him space to engage on his own terms, and the quality of our collaboration improved dramatically.
This isn’t about walking on eggshells. It’s about recognizing that feedback is a communication act, and communication is shaped by culture. A framework that ignores that isn’t a framework — it’s a script.
The Speed Problem
In an agency, feedback needs to be fast. Clients have deadlines. Campaigns have launch dates. You can’t sit on feedback for three days while you craft the perfect message.
So I’ve built a tiered system.
Tier one: Inline comments. For small tactical feedback — a word choice, a design element, a data point — I leave inline comments directly in the document or design file. No meeting needed. The person processes it asynchronously.
Tier two: Voice note. For medium-level feedback that needs context but not a full conversation, I send a two-minute voice note. It’s faster than typing and carries tone, which text doesn’t. The person can listen, process, and respond on their schedule.
Tier three: Live conversation. For significant strategic feedback, direction changes, or anything that might be emotionally loaded, we talk. Not over text. Not over email. A call. Because nuance requires real-time interaction, and the most important feedback is always the most nuanced.
The mistake I see other managers make is using tier three for tier one feedback. They schedule a thirty-minute call to say “change the headline.” That’s not thorough — it’s inefficient, and it trains the team to dread feedback conversations because they associate them with unnecessary ceremony.
The Positive Feedback Blind Spot
I’m going to be honest about a failure mode I still struggle with: I’m better at corrective feedback than positive feedback.
When something is wrong, I notice it immediately and address it. When something is excellent, I often just… move on to the next thing. The work gets approved, shipped, and I’m already thinking about the next project.
This is a problem. Because feedback is how people calibrate. If they only hear from me when something is wrong, they start to assume that silence means “good enough.” And “good enough” becomes the ceiling, because there’s no signal telling them when they’ve exceeded it.
I’ve forced myself to be more deliberate about this. When Andre designs something that makes me genuinely pause — that hits an emotional register I wasn’t expecting — I tell him. Specifically. Not “great job” but “the way you used negative space in that hero section creates a sense of calm that perfectly matches this client’s positioning as the steady, reliable option. That was a brilliant choice.”
Specific positive feedback does two things. It reinforces the specific behavior you want to see more of. And it builds the trust that makes corrective feedback land softer, because the person knows you see their excellence, not just their gaps.
What I’m Still Learning
I don’t have this figured out. Every few months, I give feedback that lands poorly — too blunt, too vague, too late. The framework doesn’t prevent mistakes. It reduces them and gives me a structure for recovering when they happen.
The biggest thing I’m still learning is patience. Feedback is not about immediate correction. It’s about gradual calibration. You’re not trying to fix this one deliverable — you’re trying to shape how someone thinks about their work over months and years. That requires repetition, consistency, and the willingness to say the same thing twelve different ways until it clicks.
The best feedback I ever received was from a mentor who told me, three separate times over six months, that I was solving the right problems with the wrong framing. The first time, I nodded politely. The second time, I started to see what he meant. The third time, it clicked, and it changed how I communicate strategy to this day.
He was patient enough to let the feedback compound. That’s the part most people skip.
Give feedback like you’re planting trees, not putting out fires. The results are slower, but they last.