Episode Details

Scot Wingo sits down with Andrew Bell, ReFiBuy's new VP of Research and host of the new RetailPlaybook channel, for a deep dive on Alexa for Shopping, Amazon's patent playbook, and what agentic commerce optimization looks like across Amazon, Walmart, and Target.

This is a launch episode. We're announcing two things at once: Andrew Bell joining ReFiBuy as VP of Research, and the debut of RetailPlaybook. Going forward Andrew hosts Retail Playbook himself, but for episode one Scot jumped in to introduce the channel and interview Andrew in person. It's airing here on Retailgentic as a crossover. 

Highlights: 

• 𝗚𝗼𝗼𝗱 𝗔𝗖𝗢 𝗶𝘀 𝗽𝗹𝗮𝘆𝗯𝗼𝗼𝗸-𝘀𝗽𝗲𝗰𝗶𝗳𝗶𝗰: "Good ACO on Amazon means optimizing for Alexa for Shopping, while also keeping the basic mechanics of SEO" — two distinct layers in symbiotic union, and a different playbook for every marketplace.

• 𝟭𝟬𝟬 𝗺𝗶𝗹𝗹𝗶𝗼𝗻 𝘂𝘀𝗲𝗿𝘀: Amazon just announced its shopping agent (the artist formerly known as Rufus, now Alexa for Shopping) is reaching roughly 100 million people.

• 𝟲𝟬% 𝗰𝗼𝗻𝘃𝗲𝗿𝘀𝗶𝗼𝗻 𝗹𝗶𝗳𝘁: Andrew is seeing conversion rates climb ~60% on product pages in particular.

• 𝗔𝗺𝗮𝘇𝗼𝗻'𝘀 𝗲𝗰𝗼𝗻𝗼𝗺𝗶𝗰 𝗲𝗻𝗴𝗶𝗻𝗲: ~$400B in counted revenue, with true GMV likely north of $600B once you account for third-party sales.

• 𝗕𝘂𝗶𝗹𝘁 𝗼𝗻 𝗖𝗹𝗮𝘂𝗱𝗲 𝗮𝗻𝗱 𝗡𝗼𝘃𝗮: Alexa for Shopping runs on Claude, Amazon's own Nova, and custom models — and the question that matters is "is your product being reasoned through?"

• 𝗧𝗵𝗲 𝗽𝗹𝗮𝘆𝗯𝗼𝗼𝗸 𝗺𝗲𝗻𝘁𝗮𝗹 𝗺𝗼𝗱𝗲𝗹: Andrew frames each marketplace like a different era of basketball, Jordan's two-pointer game vs. today's spread, zone-defense, three-point era. Different era, different playbook.

• 𝗙𝗿𝗼𝗺 𝗔𝗻𝗰𝗶𝗲𝗻𝘁 𝗚𝗿𝗲𝗲𝗸 𝘁𝗼 𝗔𝗦𝗜𝗡𝘀: Andrew studied Ancient Greek and preaching, then self-taught business by reading every book on the Barnes & Noble shelf over six months.

• 𝗣𝗮𝘁𝗲𝗻𝘁 𝗧𝘂𝗲𝘀𝗱𝗮𝘆: Andrew reads Amazon's granted patents every Tuesday (and non-granted ones Thursday), scoring each on a levels 1–5 framework because "patents are Amazon's playbook of the future."

• 𝗗𝘂𝗲𝗹𝗶𝗻𝗴 𝗨𝗖𝗣 𝘁𝗵𝗲𝗼𝗿𝗶𝗲𝘀: Andrew thinks Amazon joined the universal commerce protocol to publish its data outward; Scot thinks they'll pull UCP inward. They agree to see who's right.

A new channel, a new voice, and a host who reads Amazon patents for fun. If you sell on Amazon, Walmart, or Target, as a brand or a third-party seller, this is the playbook you'll want open. Enjoy the conversation.

Timestamps: 
01:02 Scot introduces the new podcast
02:39 In studio — Retailgentic x Retail Playbook, first time meeting IRL
03:06 Andrew joins ReFiBuy as VP of Research + channel launch
04:13 The vision for Retail Playbook (and how it differs from Retailgentic)
05:54 "ACO everywhere" — answer engines vs. retailer-hosted agents
06:09 Amazon's scale: $400B+ and 100M Rufus users
06:55 60% conversion lift; vertical + horizontal agents; built on Claude and Nova
08:34 Why playbooks work — the Jordan-vs-LeBron era model
11:48 "Christopher Nolan style" — rewinding to Andrew's story
12:02 Andrew's background: Ancient Greek, Bible school, self-taught
12:23 Touch of Class: $100K/mo to $4M, 1,500 pages by hand
13:00 GPT store, tens of thousands of sellers, 1,000 ad campaigns
15:08 NFPA: unauthorized resellers and 3x buy box
16:25 "Amazon Science Made Simple" — patents as source of truth
18:08 Patent Tuesday and the levels 1–5 framework
19:07 Use-case language and predicting the Alexa–Rufus fusion
20:51 How Alexa for Shopping ranks: 100 → 30 → the best 5–8
22:39 Full autonomy: price-trigger auto-buy
23:18 Sparky goes live on the web — and where Walmart lags
25:35 The Death Star flywheel and Amazon's selection obsession
27:34 Shop Direct and the dueling UCP theories
28:53 ChatGPT's