Think inside your AI world.
Is this channel scaling or saturating?
A saturating channel looks a lot like a scaling one from the top line: spend up, revenue up, everyone pleased. The tell is hidden underneath, in the marginal cost — whether the last pound you added cost more to convert than the one before it. Total numbers rising can mask a channel that has quietly stopped being worth feeding.
Scaling and saturating share a surface: both grow. They diverge in the marginal CAC, and that is the number nobody watches, because the dashboard leads with totals. Unl holds the saturation line you set — the marginal CAC past which more spend is destroying value — and reads the channel against it, so the verdict distinguishes real scale from expensive stagnation.
Why do the two look identical?
Because absolute growth is the default lens, and absolute growth is present in both. When you double spend on a channel and revenue rises, the instinct is to call it working — and it might be, or you might be paying steeply more for each incremental customer while the average, dragged up by cheap early conversions, still looks fine. The average hides the margin.
Saturation is a marginal phenomenon, so it is invisible to a total-based read. A channel can be simultaneously growing in aggregate and unprofitable at the edge, and only the marginal CAC — cost of the next customer, not the mean customer — reveals it. That is precisely the number the top-line dashboard is built not to show.
What does the verdict measure against?
Say you scale acquisition for a subscription app and set the line explicitly: a channel is saturating once its marginal CAC exceeds £120, because beyond that the incremental customer never clears your contribution margin. Read against it, the channel returns a verdict the top line would have hidden: “saturating — marginal CAC is £140 even though blended CAC is £95; the extra spend is buying customers you lose money on.”
The £120 marginal line and the contribution-margin reasoning are yours, not the channel’s. A model can see spend and revenue both rose; it cannot call saturation, because the marginal line that defines it is a decision about your unit economics, not a fact in the ad platform. The frame judges the data it is given; it does not verify the source’s accuracy.
What does knowing this change?
It changes the next pound. Told the channel is saturating at the margin, you hold spend at the level where marginal CAC still clears your line, rather than pushing it because the total looked healthy. And when you improve contribution margin later, you raise the £120 line in Unl, and the channel’s headroom is re-read against the new economics.
So “scaling or saturating” stops being a story told from the top line and becomes a verdict against the margin. The channel that looked like a success because everything was going up is caught at the point where up stopped meaning good.
A channel can grow spend and revenue while saturating at the margin, and total-based dashboards hide it; measured context reads the marginal CAC against the saturation line you ratified, so the verdict separates real scale from spend that is buying customers you lose money on.
Reads through Unl arrive with measured context — in the presence of the decisions you’ve already settled. The reach lane is live: one box, paste anything. If it speaks MCP, Unl can reach it. Readings arrive unprompted, the data beside the criterion; Unl is a courier, not a warehouse, and keeps only your keys and the frame.
Read further
Questions people ask
How do I know if a channel is saturating?
Watch the marginal CAC, not the totals — the cost of the next customer, not the average one. A channel can grow spend and revenue while each incremental pound costs more than it returns. Read against a saturation line you set, e.g. marginal CAC above £120, the verdict catches it even when blended CAC and the top line still look healthy.
Why does a saturating channel still look like it’s working?
Because absolute growth is present in both scaling and saturating — spend up, revenue up. Saturation is a marginal phenomenon, invisible to a total-based read: the cheap early conversions drag the average down and hide that the next customer is unprofitable. Only the marginal CAC against your line reveals it.
Can analytics tell me if I’ve hit diminishing returns?
It can show marginal cost rising if you dig for it; it can’t call saturation, because the line that defines ‘too far’ is your contribution-margin decision, not a number in the ad platform. Measured context holds that line and reads the channel against it. The reach lane is live: one box, paste anything. If it speaks MCP, Unl can reach it. Readings arrive unprompted, the data beside the criterion; Unl is a courier, not a warehouse, and keeps only your keys and the frame.
What this is
Think inside your AI world — you stay in command
Unlimitless (Unl to friends) holds what you've settled, reads what your tools are showing, and catches what's changed out in the world — and hands your AI whatever bears on the work, the moment it's needed, without you asking. The right thing, in front of the model, unprompted, with you in command of the call. So you keep moving toward what you set out to build, on top of everything you've already decided.
It plugs into Claude, Claude Code, ChatGPT and Cursor as an MCP connector. Quick to connect, in a couple of steps.
Unlimitless is open now to invited Alpha. Apply for the Beta waitlist to come in ahead of the full launch:
Alpha is invite-only · free at launch.