Creative Queue · Marketing

Creative Queue · through Nov 13, 2026

2026 10 29 ittai adrian pipeline review

date: 2026-10-29 title: "Ittai + Adrian: Creative Queue and Voice Align pipeline review, plus a used-equipment side thread" products: [creative-queue, voice-align, the-good-tracker] attendees: [Adrian, Ittai] duration: 96m grain: https://grain.com/share/recording/999840b5-b0a5-4bc6-a118-4fa8fad2bf8f/NKREqSs8bNXXIOKXkqCO6hXlxzUlKCWblhZO4ejD tags: [gtm, linkedin, reddit, x-twitter, pipeline, voice-align]

Ittai + Adrian: pipeline review (Oct 29)

Thursday, Oct 29, 2026, 2:00 PM America/New_York, 96 minutes. Ittai came in with four agenda items: a Taylor / used-equipment question, Creative Queue, Voice Align, and whether to productize AI consulting. They booked a Tuesday follow-up, which is the already-filed Nov 3 meeting.

This recording is the immediate parent of the current Creative Queue LinkedIn motion: find brands, find the people at those brands, get a way to contact them, then connect. Nov 3 then chose cold outreach / LinkedIn first and the 3–4 people per company rule.

Creative Queue

Ittai walked a Grok-bot plan whose opening line was that Creative Queue's bottleneck is real conversations with people who can pay, not more landing-page work (he tied that to a recent Brandon comment). The suggested north star was qualified-buyer calls with a founder or budget owner, and the plan judged every channel only on whether it produced those calls — and therefore paused headline tests, SEO pushes, and ads until there were conversations.

Adrian did not agree with pausing the other work. He wanted those things happening at the same time. Ittai agreed. Adrian also split "getting people on the phone who could be build partners" from "identifying qualified buyers" as maybe two motions.

What he wanted first, before copy or spend:

  1. Research that actually finds good-fit brands and people, and accumulates them somewhere they can audit. The previous Reddit round had not done a good job of finding things worth responding to.
  2. If they cannot build a reliable pipeline of people they can contact, the rest of the plan does not matter.
  3. Run another Twitter-style research pass for Creative Queue, tune it, and turn it on. Cross-reference a Notion list of people they had collected by hand.
  4. For each brand, find a group of contactable people — founder, creative director, head of marketing, or similar — with LinkedIn and/or Twitter/Instagram. Not the brand alone.
  5. He mentioned a Good Content tactic: on the order of 100 LinkedIn connection requests a month toward the target market, per client. The starting motion here is identify people and connect.
  6. A working volume he named: if something surfaces on the order of 15–20 people a day related to these brands, they could spend about two to three hours a week sending connection requests and messages. He called that a great starting position.
  7. Separate the different motions more clearly on the page so they are not mixed together.
  8. Do not spend on the other "hill" (ads / extra content spend) until they know they can pull the right people; conversations will also inform the story. Ittai still wanted solid content for cold-outreach people; Adrian agreed on the need and asked to do it in the right order.

They opened the latest Grok list together. Adrian: even looking right away, Abbreviated was a good fit; NOOWORKS ("the new works" in the transcript) also looked right; a multi-brand / portfolio player looked similar to Manhattan and was viable as a separate search pattern. Ittai: this was better than any list they had made before from an agentic pass, and the closest they had gotten to the brand avatar. His open question in the room: Manhattan and Urban Outfitters had entirely different pain points, so how do they get close enough to ask what the pain is? Adrian's answer was the people step above — once you can reach the right person, you can ask.

Next step they set: feed this conversation back in, have the bot try again (even just taking the brands already found and looking up people would be a first step), then review on Tuesday. They were testing whether the agents have enough context to keep populating who to reach out to. Ittai said Grok had been more organized than Claude or Codex for him, and better at executing follow-up once the rules were set. Adrian said he did not care which platform, only that the targeting worked.

Voice Align

They looked at a Grok pass that had surfaced vibe-coded apps asking for feedback, plus Reddit and other channels.

  • Adrian: "Royce is not a thing" — do not consider them. Stripe is already connected. He was overhauling the site and thought the pricing section needed to come back.
  • Twitter persona (people vibe-coding who want feedback) felt right. The public replies did not. They felt spammy; Adrian could not see an authentic way to insert a comment on those posts. The value was the people, not the posts. Better as a personalized DM from the Voice Feedback / Voice Align account ("I saw you were building this and wanted feedback…") than as a public reply. Reddit replies made more sense there because people were asking how to collect feedback; that is also more of an SEO / answer-back play.
  • Daily loop: scrape those people into the spark-brain codebase, then decide messaging. He also wanted a first pass at DM copy, supporting landing-page context (Voice Align as a layer for iterating with testers), and research on a second avatar — a team lead at a more established mid-size company (he used "not necessarily Linear-scale") with a possible LinkedIn DM and case-study path. Ittai added vibe-coding thought leaders / YouTube creators as a third pile to accumulate separately (affiliate or build-partner later, not mixed in).
  • Upcoming product color Adrian named: voice surveys / voice forms so teams can collect feedback without a call.
  • Mike Rock had flagged privacy-policy and terms-of-service drafts; Ittai slacked Adrian a Grok-bot draft to attach to the backlog.
  • Ittai's Reddit note, which Adrian asked him to send: Reddit works as an answer-back channel, not a launch channel; small honest answers on evergreen builder questions get indexed; the site has to catch that traffic and (in that note) agent readiness was 44/100. Adrian: they are missing explainer / blog content both for SEO and as something to send people.
  • Product Hunt, comparison pages, and case studies looked useful; directory/ad-template lists less so. He sent an App Studio video / transcript for launch order-of-operations, with a caveat that AI-platform pricing in that video is not one-to-one.
  • Come back the next week with another set of docs. Fix the subreddit error he had flagged. Do not wait for his new Voice Align context package.

Other topics

They also talked through a used-equipment / Taylor distributor idea (filed with The Good Tracker in the central repo, not repeated here) and an AI-consulting video. Adrian asked Ittai to look at the offer on adnh.info and send feedback by Loom or Voice Align.

A short personal check-in at the start is not filed here.