Why every plan includes everything

The standard way to price business software is to decide which features are hostages. You build thirty things, you put eight of them behind the middle tier and four behind the top one, and the pricing page becomes a grid of ticks whose real job is to make the cheap column look uncomfortable.

It works. It is also the reason nobody enjoys buying software. We went the other way, and this is the reasoning — including the part where it costs us money.

What we actually sell

Nucle.OS has three plans. Every module is on all three. What differs is how many people are in the workspace and how much of the four metered things you use — call recording minutes, voice minutes, email sends and contact enrichments.

Those four are the only parts of the product where using it more costs us more. Everything else is a row in Postgres. Charging you differently for a row in Postgres, depending on which page you happened to open it from, would be a pricing decision dressed up as a product one.

A feature gate is a promise that the software could do the thing, and chose not to.

The case against gating, from the buyer’s side

• It makes the trial dishonest. Most trials run on the top tier. You evaluate a product you are not about to buy, then discover which parts of your new workflow were the expensive parts.
• It punishes small teams for being small. A one-person studio has exactly the same need for e-signature and a client portal as a twenty-person one. It just has less leverage.
• It turns the product into a negotiation. Every gated feature is a support conversation, a discount request, or a churn reason, and all three cost more than the feature.

The case against gating, from our side

This is the part that is usually left out. Tiering by feature is not only a revenue tactic — it is also a way of not having to know what anything costs. If you cannot work out your unit economics, you can always put another feature behind another tier and watch the average revenue go up.

Pricing on usage forced the opposite. We had to know what a recorded minute costs, what an enrichment costs, what an email send costs, and what a workspace consumes in a normal month. That is uncomfortable early — we are still calibrating it, which is why the overage rates in the code are marked as provisional and are not switched on — but it is the only version where the price and the cost are related to each other.

What it costs us

Three things, honestly.

• The obvious upgrade path is gone. We cannot grow an account by withholding something; it grows when the team grows, or not at all.
• The top tier is harder to justify on a slide. “Everything, plus more of the things that cost money” is a worse headline than a column of extra ticks, even when it is a better deal.
• Heavy users on the cheapest plan are genuinely less profitable. We have decided we would rather have them, and would rather find out where the line is from real invoices than from a guess.

What would change our mind

If a module ever carries a real per-workspace cost — a provider we pay per seat, a licence we resell — it will be priced separately and said so plainly, in the place you would look. What will not happen is an existing, already-included module moving behind a tier. That is a bait and switch even when the terms permit it.

The plans, the allowances and what happens at the edges are on the pricing page, and the full list of what “everything” means is on product.