Turning Win-Loss Interviews Into Better Ad Creative
Most performance creative teams find out an ad stopped converting from the numbers, not from the person who almost bought and then didn't. Win-loss interviews close that gap: a short conversation with someone who saw your video or display creative and either bought or walked away, held while the decision is still fresh.
Done well, the output isn't a sentiment score. It's a short list of specific things to change in your next creative brief: on messaging, offer clarity, proof points, or format.
Vendors Covered in this Article
Disclosure: We may earn a commission if you buy through some links on this page. It doesn't change what we recommend.
Who to call, and how soon after the decision
Call people within about two weeks of the decision, closed-won and closed-lost alike, using the same rough window for both so you're not comparing a fresh loss to a stale win. If a deal involved a salesperson, get their take too: reps often remember which specific frame of a video ad or which line in an email the buyer quoted back to them, details a self-serve buyer skips over in a form.
Avoid batching these into a single research sprint once a quarter. By the time you call, most people have stopped associating their decision with a specific ad, and you'll get generic answers about price and timing instead of anything a creative team can act on.
The questions that actually explain a creative's performance
Skip broad questions like "what did you think of our marketing." Ask about the specific touchpoint:
- Which ad or video is the first one you remember seeing, and where
- What did you think it was offering, in your own words
- What almost made you stop watching or scroll past
- What would have needed to be true for the offer to feel obviously worth it
- Did anything in the ad contradict what you found once you clicked through
The second and fifth questions do the most work. A mismatch between what someone thought the ad promised and what the landing page actually delivered is one of the more common, fixable reasons a well-produced video underperforms.
Coding the transcripts so patterns show up
A handful of interviews feels anecdotal until you tag them consistently. Set up a short taxonomy before you start calling, not after: message clarity, offer clarity, trust or proof gap, format mismatch (someone expected a demo and got a testimonial reel, say), and timing or budget, which is a real reason but not a creative one.
Tag each answer against that list instead of writing a loose paragraph of notes per call. After even a handful of interviews, one or two tags will usually show up far more than the rest, and that's your signal, not any single memorable quote.
For example, a team completes several interviews and tags each answer. Most of the trust or proof gap tags come from buyers who said the ad's claim sounded too good to be true, while offer clarity tags come mostly from people who could not tell what happens after clicking. Those are different fixes: the first calls for a named customer or a specific result on screen, and the second calls for a clearer first frame that states the offer. Tagging keeps the team from treating them as one vague problem called weak creative, and it shows which change to test first in the next round.
Turning themes into a specific line in the next brief
The step teams skip is translating a theme into something a creative brief can use. "Buyers didn't trust the ROI claim" isn't a brief note. "Replace the on-screen stat card with a named customer saying the number in their own words" is.
Rewrite every theme this way before the debrief with your creative team or agency: name the specific asset, the specific change, and the specific interview quote that justifies it. If you use Martha, MeetMyCMO's AI CMO, she can turn a batch of tagged notes into a first draft of that brief update, though a person should still sanity-check the quotes before it goes out.
Where win-loss interview programs quietly fail
The most common failure is treating three or four interviews as proof of anything. Small samples are still useful for generating hypotheses, but say so out loud rather than presenting them as validated findings; a theme from four calls deserves a test, not a rewrite of the whole campaign.
The second failure is only interviewing losses. Winners tell you what worked, and that's just as actionable: if the same proof point comes up unprompted call after call, that's a reason to make it more prominent, not just a nice thing to have heard.
What Good Looks Like
Good practice means every closed-lost deal above a meaningful size gets a short interview within a couple of weeks, and every creative-relevant theme gets logged in a shared taxonomy your creative team or agency can search before the next brief.
Building The Capability (5-Stage Skill Ladder)
How to Get Started
Disclosure: We may earn a commission if you buy through some links on this page. It doesn't change what we recommend.
Frequently Asked Questions
How many win-loss interviews do we need before we trust a theme?
There's no fixed number, but treat anything under about ten interviews as a hypothesis rather than a conclusion. If the same specific reason, a confusing offer or a missing proof point, shows up in three or more calls independently, it's worth testing in the next creative round. Below that, note it and keep collecting instead of rewriting a whole campaign off a couple of calls.
Should the person running these interviews also sit on the creative team?
It helps to keep some separation. A neutral interviewer, someone from product marketing or an outside researcher, tends to get more candid answers than the person who made the ad asking whether the buyer liked it. A creative lead can still sit in and listen, just not lead the questioning.
What's the fastest way to start without dedicated research staff?
Ask sales to flag the next five closed-lost deals of a decent size, and get thirty minutes on the phone with each buyer using the same five questions every time. You don't need transcription software or a formal report; a shared doc with tagged notes from those calls will already tell you more than the win-loss data you have now.
About the numbers
This guide doesn't quote a sourced benchmark. Figures in it are estimates or general guidance, so check them against your own numbers.
Related Guides
How to Run Win-Loss Interviews That Change Your Next Pitch
A practical guide to interviewing closed-won and closed-lost prospects: who to call, what to ask, and how to turn the answers into a better pitch.
How to Run Win-Loss Interviews Sales Will Trust
A practical way to schedule, question, and code win-loss interviews so the findings change what sales and marketing actually do next quarter.
Win-Loss Interviews: What to Ask When Trials Go Cold
A guide to picking who to interview after a trial ends, the questions that surface real objections, and how to turn the answers into onboarding fixes.
How to Turn Win-Loss Interviews Into a Working Feedback Loop
A practical way to run win-loss customer interviews, sort the real reasons deals close or die, and route findings to the team that can fix them.
What Win-Loss Interviews Really Tell You About ABM
Who to interview after a deal closes, which questions get honest answers, and how to turn win-loss patterns into changes to your ABM targeting and messaging.
How to Turn Win-Loss Interviews Into a Retention Email Program
A runbook for running win-loss customer interviews and routing what you learn into onboarding and retention emails, not just a slide deck.