Positioning, packaging, format design, production system and honest measurement for a channel that compounds.
YouTube Video Ideas
You are building a complete YouTube strategy. The platform rewards consistency and packaging above production quality, and most channels fail by inverting that.
## Brief
- Company:
- Product: A project management tool for creative agencies
- Audience: Heads of marketing at 20–200 person B2B SaaS companies
- Problem solved: Scope creep going unnoticed until the project is unprofitable
- Proof available: 31% reduction in write-offs within two quarters
- Desired action: Start a 14-day free trial
- Capacity: One person, one video a week
- Voice: Direct, practical, dry humour; never hypey
---
## Phase 1 — Is YouTube right, and at what cost?
Video is the most expensive content format per unit. Assess honestly:
- Does Heads of marketing at 20–200 person B2B SaaS companies search for or watch content about Scope creep going unnoticed until the project is unprofitable?
- Can One person, one video a week sustain a genuine cadence? A channel publishing sporadically does not build.
- Is there a route from viewing to Start a 14-day free trial? YouTube is poor at driving immediate clicks, so it suits consideration and trust more than direct response.
- Does Northpath — workflow tools for creative teams have someone who can be on camera, or is the content demonstrable without a presenter?
If the honest answer is that another channel is a better use of One person, one video a week, say so.
**CHECKPOINT 1** — Present the assessment. Stop.
---
## Phase 2 — Channel positioning
- What is this channel *about*? Specific enough that a viewer knows what they will get.
- What recurring point of view does it hold?
- Who is it not for?
- What does a subscriber get, stated as a promise?
YouTube rewards narrowness more than most platforms — the recommendation system needs to understand who to show your videos to, and a channel covering everything gets shown to nobody.
---
## Phase 3 — Format design
Design three repeatable formats rather than making one-off videos. Repeatable formats reduce production cost, build audience expectation, and let the algorithm learn the audience.
For each: the structure, the typical length, the production requirement, and the job it does — search, browse, or conversion.
Balance:
- **Search-led** — steady long-term views, matches specific queries
- **Browse-led** — bigger spikes, depends on packaging
- **Conversion-led** — smaller audience, closer to Start a 14-day free trial
---
## Phase 4 — Packaging system
## Packaging decides everything
On YouTube, title and thumbnail determine whether a video is watched. A strong video with weak packaging is invisible; the platform never surfaces it long enough to find out it was good.
Treat title and thumbnail as the primary creative work, not as something done after the edit. Design them **before** scripting — if you cannot package the idea compellingly, the idea itself is the problem.
Establish the process:
- Title and thumbnail designed **before** production
- The idea is validated by whether it packages well — if it does not, it is not made
- A consistent visual language so the channel is recognisable in a feed
- The specific test: is the thumbnail legible and comprehensible at small size?
Then produce ten packaged video ideas — title and thumbnail description — that fit the positioning and formats.
---
## Phase 5 — Production system
- **Batching** — how to produce several videos per session, since setup is the dominant cost
- **The minimum viable production quality** for this audience. Over-production is a common cause of channel death: the standard becomes unsustainable and publishing stops.
- **The workflow** from idea to published, and where it typically stalls
- **The quality bar** — a falsifiable test each video must pass
Then write a **full script for the strongest of the ten ideas**, with retention design.
**CHECKPOINT 2** — Present formats, packaging and production system. Stop.
---
## Phase 6 — Measurement and expectations
- **The metrics that matter**: click-through rate on impressions, and average view duration. These two drive everything the algorithm does.
- **How to read them**: low CTR is a packaging problem; high CTR with low retention is an over-promising problem; both low means the idea was wrong.
- **The realistic timeline** — YouTube is slow and non-linear. State it honestly so the channel is not abandoned at video eight, which is where most channels die.
- **The counter-metric**: is the channel attracting Heads of marketing at 20–200 person B2B SaaS companies or a broader audience that never converts?
- **The stopping rule.**
---
## Phase 7 — The honest close
- What this achieves in twelve months at One person, one video a week
- The most likely failure cause — usually inconsistency or unsustainable production standards
- The single highest-leverage change if only one were possible
---
## Standing rules
- Never invent view counts, benchmarks or channel results. Use [NEEDS DATA].
- Disclose paid promotion and sponsorships in-platform.
- Never recommend misleading packaging. Clickbait that under-delivers is punished by the retention signal that actually drives the platform.
- If Northpath — workflow tools for creative teams cannot sustain a cadence, say plainly that they should not start.
## Never fabricate
Do not invent statistics, customer names, quotes, case-study numbers, testimonials, or research findings. If you need a figure you have not been given, write [NEEDS DATA] and say what you need. Realistic-sounding invented numbers are the fastest way to destroy credibility with an informed audience.
## Avoid
- Opening with a definition, a rhetorical question, or "In today's fast-paced world".
- The words "delve", "tapestry", "landscape", "game-changer", "unlock", "elevate", "seamless", "robust", and the construction "it's not just X, it's Y".
- Padding: any sentence that restates the previous one in different words.
- Evenly-weighted sections. Real arguments have a centre of gravity.
- Hedging every claim into meaninglessness. Take a position.