A client walked me through a pricing change last week that they were in the middle of losing an argument about.
The company sells a self-serve digital product. One flagship offering, about $300, sold for years as the premium thing they make. They had just launched a genuinely better version with AI built into it, and the plan was to move that new version into the $300 slot and drop the original to around $80 — down into the territory where the discount competitors live. The ones who had been quietly eating their market for eighteen months.
Their co-founder hated it, and you can see why. That $300 product had been the flagship since the beginning. Cutting it by roughly seventy percent reads as a markdown of the thing the company is proudest of, and the spreadsheet agrees: same product, a fraction of the revenue per sale.
They were both reading the number off the wrong ledger.
Two readings of the same price cut
Read as a product decision, dropping the original from $300 to $80 is a seventy-percent revenue cut on your best-known offering, and you only do it because the market forced you to. It is a retreat with a press release attached.
Read as an acquisition decision, the identical move says something else: we are going to spend about $220 a head to acquire a paying, category-educated customer who has handed over their email and their card, who will be inside the product for months, and who now has a specific, provable reason to care about the $300 version we would actually like to sell them.
Nothing about the decision changed. Only the ledger it gets posted to. I said a version of this on the call — you are taking a hit on price, but what you are buying is attention — and they wrote it down, because they were walking into the room with that co-founder the next morning.
Why “lost revenue” is the wrong comparison
The comparison that actually decides this is not $300 against $80. It is $220 against whatever it currently costs you to put a stranger in front of your premium product and get them to seriously consider it. For most companies under $10M ARR that number is not small, it is not stable month to month, and what it buys you is a lead rather than a customer.
A customer of the cheap tier is a materially better asset than a lead of the same cost, and it is worth being specific about why rather than waving at it.
- They paid. Not a form fill — a transaction. They have a budget line for this category and they have already cleared the hardest psychological hurdle there is, which is the first one.
- They are using the thing. A lead has an opinion about your product. An $80 customer has an experience of it. That turns the upgrade conversation from whether into how much more.
- They self-selected on the problem. Nobody buys a bookkeeping tool, an inventory tracker, or a compliance dashboard by accident. The qualification happened before you spent anything.
- The clock is working for you. My client’s entry product comes with a year of access, which means every single one of those customers walks into a renewal moment on a known date. That is not a one-shot channel. It is an owned list with a scheduled upgrade conversation built into it.
That last point is the one most teams miss when they model this. They treat the entry tier as a single discounted sale instead of as the top of a funnel they now own outright and do not have to rent from an ad platform every month.
Like this? I send one idea like it most weeks. One PLG idea a week → subscribe
Three conditions, or it is just a discount
Reframing a price cut does not make it a good idea. This move works under three conditions and it fails loudly without them, so check all three before you touch the price.
1. The upgrade has to be visible from inside the cheap tier
If the only place your premium tier exists is the pricing page, you bought the attention and then threw it in the bin. The $80 customer should run into the absence of the premium version in the normal course of using what they bought — not as a nag, but as a legible gap. Showing someone that value exists and charging them to resolve it is a far stronger position than hiding the feature entirely, because they can size the prize before they decide. That is the same logic that governs where a paywall belongs, which I worked through in The Math Behind Your Value Gate.
2. The tiers have to differ on something the user can feel
My client’s split is unusually clean. The basic product is static material you work through on your own. The premium one is an interactive AI layer over the same material that responds to what you do. That is not a feature-matrix difference, it is a different experience of the same job — which is exactly what makes the upgrade ask land.
When the two tiers differ by seat count, or by a checklist of capabilities nobody can picture, the entry tier does the opposite of its job: it teaches the customer that your product is unremarkable, and they leave with that as their only impression. You paid $220 to install a bad opinion.
3. You have to survive the mix shift
A working version of this move looks, at first, like a disaster. Many more customers paying much less, immediately, with the upgrade revenue arriving several months later. If the entry tier is cash-flow negative per customer, you have bought a marketing channel you cannot afford to run.
There is a second leak to check before you publish the number. A deep public price on your bottom tier becomes a data point inside every larger negotiation you have open, and the ceiling you can defend at the top gets set by the floor you advertise at the bottom. That is one of the three gates I use on any proposed discount, and it kills more of them than the margin math does — the full set is in Three Dollars Is Not Why They Didn’t Buy.
What this is not
It is not a discount. A discount is the same product at a lower price for a reason the buyer reads as temporary, and it trains them to wait. This is a permanent rung with a different job from the one above it, priced on purpose and held there.
It is not freemium either, though it is in the same family. Freemium gives the entry rung away and relies on in-product behavior to do the qualifying, which means you need volume and a product that proves itself without a human in the room. A paid entry tier qualifies harder and yields fewer people, and the choice between them comes down to how much traffic you actually have and whether the product can land its First Value Moment unassisted. The four questions that settle it are in Free Trial or Freemium?
The rule
If your price does not vary by segment, rank your segments by what they cost to reach rather than by how good they would look on a logo wall. Then apply the same thinking one rung down your own ladder.
Here is the question to put to it: what would your entry tier be worth if it produced no profit at all and produced only upgrades? If the honest answer is “quite a lot,” then you have been running an acquisition channel and reporting it as a product line — and defending its margin is costing you the customers it exists to go and get.
Deciding what each rung of the ladder is actually for — and then placing the gate where the user can see what they are buying — is Diagnose and Build work in the Trial Conversion Engine.
Running a trial you’re not happy with? Book a call and I’ll walk your funnel live.
