---

date: 2026-01-30 title: "Roles at CQ: studio model, Voice Feedback as the near-term ads motion, Creative Queue later" products: [creative-queue, voice-align] attendees: [Adrian, Ittai, Brandon] duration: 34m grain: https://grain.com/share/recording/a750072c-6119-4cea-8019-36f0e37f9bc8/fJSKimqwfwhNgR9mIvpCTfwlKNXDZyNCdF6JYELc tags: [gtm, operating-model, roles, paid-ads]

---

Roles at CQ

Friday, Jan 30, 2026, 3:45 PM America/New_York, 34 minutes.

This was a Spark operating conversation tagged as "Roles at CQ." Financial split details stayed as a follow-up document Ittai would write; they are not treated as decided here.

Operating-model realization

Adrian: they have been pretending they are not a studio. They are a studio, and they should sell like one. Easier-to-sell apps can support the ones that are harder to sell but can be priced higher. He connected that to recent writing about the Spark story versus the Creative Queue story.

Brandon: looking back it is easy to see why they sequenced things this way, but they should be testing and learning instead of perfecting something in a vacuum that nobody touches.

What that meant for go-to-market

• Voice Feedback (and a separate "selects" motion Adrian named) can run ads in the background, independently of getting Creative Queue off the ground. Creative Queue, in the short term, is "not going to be an ads game."
• Brandon: Voice Feedback can be marketed now; they would get emails from interested people. Adrian: it is a few reps away from being sellable, and if they play it right they could be making recurring revenue from it on a short horizon (he said about two weeks). That made Creative Queue feel, to him, like not the most urgent vertical in that moment.
• Ittai's strongest current flow, in his own words: get clear on what the app does, put up a landing page, run paid ads, and learn from conversions. Longer-play Reddit and LinkedIn work sat outside his wheelhouse; he was willing to research how to learn it.
• Adrian's challenge: Reddit findability matters for every property, not only Creative Queue — including how people discover answers inside tools like ChatGPT. It needs a real strategy (how they show up, how they drive somewhere), not a one-off post. Brandon's simpler take: search the relevant threads and enter the conversation. Ittai treated Reddit as a longer play, more like SEO, after a paid motion was running.
• Landing pages are not enough: they need Stripe (or equivalent) so someone can actually pay.

Ittai said a small paid Creative Queue test in an earlier period had produced meetings; that earlier point is in the Jan 16 prep. This call did not reopen a Creative Queue ads decision.

Roles, proposals, and practice

Adrian asked Ittai to stop waiting reactively and to make strong proposals — a two-day turnaround, then feedback — the way he and Brandon already do on product. He named three pieces of the marketing job: positioning (including the five-word "what does it do?"), the plan of attack (paid, Reddit, and the rest), and execution.

Ittai's own summary of next steps: (1) get clarity among the three of them on how percentages and participation should feel, (2) run his landing-page-and-ads flow, (3) then start the Reddit work. Adrian's challenge matching that: Ittai writes down how the split should work, they meet and triage it, and Ittai practices making the call rather than only reacting. Brandon asked to be looped into the CSV-level detail of what ads are and are not working so they can iterate landing pages.

They also talked about needing a shared go-to-market playbook at the Spark level (inspired by a video of a four-person studio that splits work evenly and runs several apps), not only an org chart. Adrian asked Brandon to pull the playbook out of that video so they could write their own version and map each product onto a sequence.

A side thread about other collaborators and about Ittai winding down a separate engagement is not filed here. Adrian's other challenge to Ittai — get more comfortable prototyping with AI tools so ideas can hit the table faster — is noted as an operating ask, not as a Creative Queue page update.

First client

Ittai mentioned being glad they had gotten the first client. This recording does not name which client that was.