Pay
Price selects for the buyer, not just the budget
The people who pay least reliably ask for most, and this is not a coincidence.
Guides on Pay: Pricing, from first principles to the annual review, The ceiling is hours, and it arrives sooner than people expect, What rating work actually pays
The cheapest tier attracts the hardest buyers because a low price selects for them: people who compared only on price, who lose little by spending your time, or who are testing whether the category is worth anything. The result is more messages, more revisions and weaker reviews per fee earned.
This is adverse selection, and it is a property of the price rather than a run of bad luck. The term comes from economics: the 2001 Nobel economics prize announcement credits George Akerlof with showing how a market where one side knows more than the other "can contract into an adverse selection of low-quality products." Here the information gap runs the other way - the buyer knows how demanding they will be, and you do not.
Why the bottom tier selects the way it does
Three separate mechanisms push in the same direction.
The first is expectation mismatch. A buyer paying the lowest advertised price frequently has no sense of what that price buys, because the price was the only thing they compared. They arrive with the expectations set by your top tier and the budget of your bottom one, and the gap becomes a revision request.
The second is opportunity cost on their side. Someone spending a small amount loses little by spending an hour of your time getting it right, so they spend it. The same person spending five times as much is usually more decisive, not less, because a considered purchase gets considered once.
The third is that a low price attracts people who are testing whether the whole category is worth anything. They are not buying your work; they are buying a sample of the market, and a sample is judged harshly by definition.
None of this makes them bad people. It makes the tier a filter that is currently set to admit exactly the buyers your process handles worst.
The evidence is in your own records
You do not need a study for this, and you should not trust one about someone else's market anyway. You need thirty jobs and four columns.
| Column | What to log |
|---|---|
| Tier bought | which line of the card |
| Messages before booking | count, not a feeling |
| Minutes actually spent | including the messages |
| Outcome | review left, score, any revision or refund |
Thirty rows is enough to see whether the pattern is real for you. Divide take-home by minutes per tier and the bottom row frequently comes out worse than the tier above it despite looking like easy work, which is the same effect tracking take-home per format surfaces from the other direction.
Be honest about the messages. Pre-sale conversation is the cost that hides most successfully, because it does not feel like working.
What to do that is not deleting the tier
Deleting it is the obvious move and usually the wrong one. The bottom tier is doing real work: it catches buyers who genuinely have a small job, it gives your middle tier something to be compared against, and it is the entry point for repeat buyers who later buy up.
Four adjustments are better.
Make the scope of the cheap tier extremely specific and extremely small. Not "a short assessment" but a named length, a named number of axes, no revision, one file. Ambiguity at the bottom of the card is where the cost lives, and a scope written as a list rather than as prose is one you can point at later - the case for a scope box is mostly about this moment.
Remove the negotiable surface. No revisions included, stated plainly and without apology. No pre-sale consultation beyond one clarifying exchange. This is not meanness; it is what makes the price possible.
Raise it until it is uncomfortable. Most bottom tiers are set by nerves rather than arithmetic. If the tier below your main one is not at least sixty percent of it, the gap is doing nothing except advertising a discount.
Make the middle tier the obvious answer. When the cheap tier is visibly minimal and the middle one is visibly complete, most buyers choose correctly on their own, which is the entire mechanism behind a three-tier card.
The reviews problem is the expensive part
The lost fee on a bad cheap job is trivial. The review is not, because it sits on the profile that sells your expensive work, and buyers reading it have no idea which tier produced it.
This is the strongest argument for either raising the bottom tier or scoping it until it cannot go wrong. A one-star review earned at the smallest price on your card discounts every price on your card.
What this looks like from the buyer's side
Buyers are not conspiring. Most of them do not know that the cheapest listing is a different product; they read it as the same work with less of something unspecified removed. The etiquette written from the commissioning side, including what buyers are told to expect when they book at the low end, is a useful corrective to the assumption that they are being difficult on purpose.
It also helps to know what has already happened to the true floor of this market. Anything generic and instant is now free, so a cheap human tier is not competing with a cheaper human; it is competing with the free automated tools, and the only thing that beats them is being specific about something a tool cannot see. Where precision is the product rather than the impression of one, the documented measurement conventions are the reference to be precise against.
None of this is an argument for matching somebody else's floor, which is a separate decision worked through in competing with someone cheaper.
Set the bottom tier at a price you would be content to receive on a bad day, scope it so a bad day cannot happen, and let it filter. The platform-side view of how tiers are presented to buyers who are choosing is on Rate Cock's judges page.