Marketecture: Get Smart. Fast.

Episode: Playwire with Jayson Dubin: Human Intelligence vs Machine Learning in AdTech at Marketecture Live
Date: December 29, 2025
Host: Ari Paparo
Guest: Jayson Dubin (CEO & Founder, Playwire)

Episode Overview

This episode features a live interview at Marketecture Live with Jayson Dubin, CEO and founder of Playwire. The conversation focuses on the interplay between human intelligence and machine learning in the adtech sector, highlighting how Playwire leverages both to maximize publisher revenue and operational efficiency. The dialogue dives into Playwire’s development of hybrid AI-human solutions for traffic shaping, price flooring, supply path optimization, and the creation of their new platform, RAMP.

Key Discussion Points & Insights

1. Playwire’s Mission & Approach (02:03–04:06)

• Jayson Dubin introduces Playwire: A revenue and efficiency optimization technology and services company for publishers, supporting about 1,000 websites with a small team (~150 people) through automation and AI.

• On the “false dichotomy” of human vs. machine: Jayson posits publishers don’t need to choose between pure automation and pure human involvement. The most robust solutions entwine both.

  “There’s a false dichotomy going on. Publishers feel they need to go either one way or the other… But the reality is you need both of them to… figure out these complex problems."
  —Jayson Dubin [03:05]

2. Where AI Excels: Traffic Shaping & Price Flooring (04:07–08:36)

• Traffic shaping:

  • Definition: Feeding exchanges and platforms (SSPs/DSPs) the precise inventory they prefer, increasing signal fidelity and purchase rates.
  • Impact:
    • Control group (no AI): +9% Revenue Per Session (RPS)
    • Shaped by AI: +21% RPS (net +12% delta)
  • Secondary benefit: 17% reduction in bid requests per session led to faster page loads and higher site traffic.

  “By feeding them more of that [preferred inventory], they're going to get better signal fidelity and they're going to end up buying more of our inventory… On the experiment where we did have AI, we saw a 21% increase in RPS.”
  —Jayson Dubin [04:35–05:25]

• Price floor controller (PFC):

  • AI outperforms manual set-up: By combining nine variable dimensions (e.g., ad size, time, refresh count) and deploying real-time adjustments, they average 1.2 million price floors per site per day.
  • 20% RPM (Revenue Per Mille) uplift.

  “We average 1.2 million price floors per site, per day… again, you see a 20% uplift in RPS.”
  —Jayson Dubin [06:38]

3. Human Input Drives Strategy & Context (03:19–04:00, 07:24–08:10)

• AI limitations: While data can identify problems, contextual and strategic responses require human judgment.

  “Data could tell you there's a problem and they can tell you something's not working. But what they can't tell you is what's the answer to that problem? That's where contextual thinking comes in.”
  —Jayson Dubin [03:51]

• Origin of innovations: Many breakthroughs, like focusing on quality over quantity, came from conversations with industry partners, not from data alone.

4. Quality, Performance, Transparency (QPT) Initiative (07:25–09:22)

• Industry shift: Moved from prioritizing scale (“more bids, more ads”) to emphasizing transparency and quality.

• Actions taken:

  • Reduced SSPs from 26 to 16; aiming for 13.
  • Removed reseller lines that didn’t add value.
  • Eliminated MFA (Made For Advertising) and low-quality content partners.

• Results from QPT and AI/automation:

  • 61% reduction in bid requests
  • 58% revenue increase
  • 168% rise in CPM
  • 107% improvement in viewability
  • 8% more overall site traffic due to faster loading, better scores

  “This was a culmination of two years of work… Revenue went up by 58%. CPM soared to 168%. Viewability went up 107%.”
  —Jayson Dubin [09:13]

5. Productization: Launch of RAMP Platform (09:23–10:47)

• RAMP: Revenue Amplification Management Platform—Playwire’s new enterprise solution announced at the conference, putting control (manual or AI-driven) directly in publishers’ hands via a real-time, rules-based system.

  “Playwire has given all this technology… into one unified platform made for enterprise level publishers called RAMP… You decide whether to take the wheel.”
  —Jayson Dubin [10:25]

Fireside Q&A with Ari Paparo (11:20–21:40)

The Publisher’s Dilemma in an AI World (11:29–12:54)

• AI’s impact on publishers: Avg. 15% decrease in SEO traffic, some up to 40%.

• Top challenges:

  • Growing complexity (identity, privacy, analytics).
  • Transparency imbalance: Buy side (agencies, DSPs) sees all; sell side (publishers) gets little visibility into buyers or the value chain.

  “From our perspective, there's little transparency on the buy side to the… sell side...Why is, why don't we have gross bid value passed all the way so we can see who's taking what out…?”
  —Jayson Dubin [13:02]

• Ongoing issue of malicious ads: Difficulty in tracing origins due to bid stream opacity; resource-intensive to solve.

  “We spend hundreds of thousands… trying to determine and, or how to stop these malicious ads. It's a huge problem.”
  —Jayson Dubin [13:35]

Where Human Judgment Directs AI (15:03–16:17)

• Decision process: Human insight defines parameters, pilots, and evaluates new AI-driven tests before network-wide rollout.

• Caveat on AI adoption: “It's not this magical thing you just sprinkle on… these are complex things… It's not so easy.”

  —Jayson Dubin [15:52]

• Risks of “blind” AI deployment: Possible to create more problems without clear objectives.

Timeout Management Optimization (16:58–17:33)

• AI for publisher-SSP timeouts: Adjusted per SSP to eke out incremental yield increases.

  “Timeouts per SSP was actually something that drove results.”
  —Jayson Dubin [17:12]

AI-Generated Publisher Content (17:35–19:38)

• Dubin’s stance: Doesn’t oppose AI-generated content, but warns quality and originality remain essential as search engines get better at filtering low-value copycat content.

  “If you're not making great quality content… at some point it's going to be exposed.”
  —Jayson Dubin [18:39]

• Old model dying: The “franchise” template content approach (clone sites, duplicate content) used to work; now, Google’s crackdown devalues it.

Search & LLMs (21:00–21:14)

• Referrals from AI: Referral traffic from chatbots/LLMs like ChatGPT is on the rise—something to be embraced rather than fought.

Market Power and Scale (20:34–21:40)

• Discussion of industry power dynamics: Larger publishers (“People”/DotDash, Twitch, YouTube) wield disproportionate influence; scale leads to preferential treatment.

• Role of companies like Playwire: Advocate for smaller publishers, “like the Lorax: we speak for the trees.”

  “If we're not making more revenue for our publishers, if we're not lowering their cost to do business, then we don't deserve to be in business… publishers need companies like us to really… speak to the trees.”
  —Jayson Dubin [21:09, 21:37]

Notable Quotes

• “There’s a false dichotomy… The reality is you need both [human brains and machine learning] to figure out complex problems.” (Jayson Dubin, 03:05)
• “Data could tell you there’s a problem… But what they can’t tell you is what’s the answer to that problem. That’s where contextual thinking comes in.” (Jayson Dubin, 03:51)
• “We average 1.2 million price floors per site, per day… see a 20% uplift in RPS.” (Jayson Dubin, 06:38)
• “It’s not this magical thing you just sprinkle on… These are complex things… It’s not so easy.” (Jayson Dubin, 15:52)
• “The bigger you are, the more you can get away with… If you have scale, you get away with a lot more than someone who does not have scale.” (Jayson Dubin, 20:49)
• “We speak for the trees.” (Jayson Dubin, 21:37)

Key Timestamps for Important Segments

• 02:03 — Jayson Dubin introduces Playwire and outlines the human vs. machine learning theme
• 04:07 — Explains traffic shaping and concrete business impact
• 06:38 — Details on AI-driven price floor control
• 07:25 — Shift from quantity to quality in adtech (QPT initiative)
• 09:13 — Presents metrics/results of AI plus QPT
• 10:25 — Launch of Playwire’s RAMP platform
• 11:29–12:54 — Publisher struggles with AI, complexity, and lack of transparency
• 13:35 — The problem of tracing malicious ads through the ad tech supply chain
• 15:03 — Discussion: how Playwire decides when to deploy AI
• 17:06 — AI optimization for SSP timeouts
• 17:35–19:38 — AI-generated content and consequences for publisher value
• 20:12 — LLMs (ChatGPT etc.) as new traffic sources
• 20:48 — Discussion of scale in the industry, market power disparities

Episode Tone & Takeaways

The episode is a candid, pragmatic exploration of what works—and what doesn’t—when merging human acumen with AI in adtech. Dubin’s approach is both hands-on and humble, repeatedly acknowledging the irreplaceable value of strategy, testing, and transparency. The message: In a world of growing automation, publishers and their partners must continuously adapt, keep the “human loop” in the machine, and above all, focus on real value creation.