# Marcus Devereaux, VP of Media Operations at Clearline Media Group — read of ProxyGeo, June 8, 2026

> 11 years in ad ops and media buying, currently managing campaign verification for 23 agency clients across CPG and retail verticals. We spend about $400K/month in managed media.

## How I got here

Someone in the Ops Efficiency Slack channel (not ours, a cross-agency group I'm in) posted a link with the message "anyone tried this?" No context. I clicked it on the train to the office because we've been having a bad quarter with placement discrepancies on a Southeast Asia campaign and our current vendor, IAS, is giving us logs that don't match what the client's own pixel is reporting. I was not in buy mode. I was in "let me see if this is real" mode.

## What I clicked first

The hero line "See exactly what users see. No VPNs. No data center proxies. Real traffic." made me stop for half a second because that's the actual problem we have. Data center proxy IPs getting filtered by publisher-side fraud detection is a real issue I've complained about internally. So I kept reading.

But then they lead with the stat: "Ad fraud costs advertisers $81 billion per year." That number has been in every deck I've seen since 2019. It's the equivalent of "did you know 70% of emails go unread?" at this point. When I see it, I subtract credibility points immediately.

## Where I paused

The section labeled "Indistinguishable from Users" gave me genuine pause, but not in the way they intended. "Ad networks cannot tell the difference between verification traffic and genuine visitors." That is either the most important sentence on the page or the thing that gets this service on a publisher blocklist in six months. I've watched this exact dynamic play out with other proxy verification tools. Publishers catch on, they update their bot detection, the tool's data goes stale. There is nothing on this page that explains how they stay ahead of that arms race. Nothing about fleet rotation, IP refresh rates, or how they handle fingerprinting countermeasures. For someone who has been burned by this cycle before, that omission is loud.

## What I distrusted

Two things, and one of them is big.

First: zero named customers, zero anonymized case studies, zero "an agency with $X in managed spend verified Y placements and found Z." The page uses the phrase "Agencies, publishers, and verification firms use ProxyGeo to audit campaigns" in present tense, and then further down on the very same page there's a disclosure that says "Honest disclosure: we don't have live customers on this idea yet." Those two things cannot both be true. One of them is a lie, and I know which one.

Second, and this is the thing that genuinely broke my trust in the whole page: what is "The Wishdeal Factory"? Why is there a section on this product page that says "Adopt this idea. Browse free. Unlock for $5. Adopt for $99"? I read this three times. Is ProxyGeo a product I can buy a subscription to, or is it a business idea being sold to someone who wants to build it? Because those are completely different things and I cannot tell which one is true from this page. If I'm a buyer, I want software. If this is an idea marketplace, I am on the wrong page and the wrong audience entirely. This meta-layer completely broke any trust the FAQ section had built up.

## What would convince me

Show me one verified catch. Literally one. A campaign in Germany where the geo claim said Frankfurt and your tool came back with screenshots from a Tier 3 market in Eastern Europe and network waterfall data showing a redirected ad server. That is the product demo I want to see. Not a diagram of how it works. An actual documented placement discrepancy with the before and after. If you have that and you're not leading with it, you have a marketing problem. If you don't have it, come back when you do.

I also need to understand the IP pool churn rate. How often do residential IPs rotate? What country do you have the strongest density in? If you tell me 150+ countries but you have 12 IPs in Vietnam I need to know that before I run a Southeast Asia campaign audit through your platform.

## What I'd ask in an email reply

1. The page says you have no live customers yet. Are you currently in beta with any agencies, or is this literally pre-customer? I want to know if I'd be the first person stress-testing this on a real campaign.

2. When a publisher or ad network updates their bot detection and starts flagging your residential IPs, what is your SLA for getting clean IPs back into rotation for that market? Who's responsible for maintaining the fleet, and how do I know I'm not getting stale data?

3. The pricing is $0.49/verification on Starter. What counts as one verification -- is that a single page load at a single IP, or does a 5-country geo check on one URL count as one or five? I need to know the unit economics before I could ever pitch this internally.

## Verdict: on-the-fence

The core idea is real and the FAQ is actually more specific than most products at this stage. But the page is serving two audiences (potential buyers and potential business-idea adopters) and it is failing both of them. I cannot tell if I'm looking at a product or a pitch deck.

---
*Memo by skeptic persona, generated 2026-06-08. Studio breaks own self-grading loop.*