Tiered Offers for Solo Businesses: How to Design Good-Better-Best Packages That Sell

TL;DR. A good-better-best menu is three versions of one offer, priced so that most buyers pick the tier you designed to be picked. Platforms built for solo sellers run on it: Fiverr’s annual report for 2025 says buyers purchase packages “tiered as Basic, Standard and Premium”, and Shopify reports that most merchants are on its Basic and Grow plans while most sales volume comes from Shopify Plus and enterprise merchants. The evidence behind tiers is uneven. The pull of the middle option holds up: a meta-analysis of 142 experimental observations found middle options “significantly more often selected”, though price-quality trade-offs tend to weaken it. The decoy effect holds up poorly: Frederick, Lee and Baskin found no significant attraction effect in 27 studies where at least one attribute could be directly experienced, and a subscription decoy that lifted a target from 43% to 72% in a classroom of 59 students went from 25% to 21% when retested with 2,003 respondents. The CEO move: treat the menu as a portfolio, choose the target tier, and price every tier on contribution per delivery hour. The student move: run tier changes as experiments and build on the effects that replicate.

For choosing the underlying model, start with freelance pricing models compared. For scoping a fixed-price package before you tier it, see the productized service pricing framework. If the tiers come with a price rise, pair this with the repricing playbook.

This article is general information about pricing design. It is not personalised financial, tax or legal advice.

What is a good-better-best offer, and why do solo businesses use it?

A good-better-best offer presents one service or product at three levels of scope, speed, access or support, each at a fixed price. Buyers compare your tiers with each other instead of with a competitor’s price. The menu does three jobs:

1. It segments without negotiation. Buyers who want more say so by choosing a higher tier, not by asking for extras mid-project.
2. It sets reference points. Each price is judged against its neighbours.
3. It reveals demand. The share of buyers choosing each tier is information a single offer never gives you.

Fiverr’s annual report describes the mechanics plainly: packages vary “such as different word counts for a translation, video lengths for a video edit, or number of revisions for a logo design.” What varies is a countable quantity the buyer understands at a glance. That is the first design rule, and it comes from a regulatory filing rather than pricing folklore.

The CEO lens: a menu is a portfolio. Each tier has a price, a delivery cost and a margin, and the mix buyers choose decides your income and your calendar. The student lens: every launch and every change is an experiment, provided you record the results.

What does the research say about why tiered menus work?

Tiered menus are usually justified with five behavioural effects. They do not deserve equal confidence.

Table 1. Behavioural effects behind tiered menus, with replication status. CEOtudent analysis of the studies listed; figures as printed.

Effect	What a menu relies on it for	Key evidence, as printed	Replication status	Design implication
Compromise effect (extremeness aversion)	Steering buyers to the middle tier	Simonson (1989): brands “gain share when they become compromise alternatives”, more so when buyers expect to justify the choice to others. Simonson and Tversky (1992): two cameras split 50/50; adding a pricier third model moved shares to 22/57/21, lifting the middle model’s share relative to the cheaper one from 50% to 72%	Neumann, Böckenholt and Sinha (2016), 142 observations: robust, but the size varies up to three times the average with method; price-quality trade-offs tend to reduce it	Use it, but do not expect it to carry a menu that differs mainly on price and quality
Attraction (decoy) effect	An inferior option that makes the target look better	Kivetz, Netzer and Srinivasan (2004), as reported by Frederick and colleagues: a subscription decoy raised the target’s share from 43% to 72%	Frederick, Lee and Baskin (2014), 38 studies: significant in four of five highly abstract numeric cases, none in 27 studies with an experienced attribute; 25% to 21% in a 2,003-person retest. Yang and Lynn (2014): 11 of 91 attempts (12.1%) significant. Huber, Payne and Puto (2014) reply that it “does replicate” and has “predictable and important boundaries”	Contested. Do not build revenue plans on a decoy
Choice overload	Keeping the menu to three tiers	Iyengar and Lepper (2000): 60% of passing shoppers stopped at a 24-jam display and 40% at a 6-jam one, but nearly 30% of those who stopped at the small display bought, against 3% at the large one	Scheibehenne, Greifeneder and Todd (2010), 63 conditions from 50 experiments, 5,036 participants: mean effect “virtually zero” (D = 0.02). Chernev, Böckenholt and Goodman (2015), 99 observations, 7,202 participants: significant once complexity, task difficulty, preference uncertainty and an effort-minimising goal are accounted for	Conditional. Complexity matters more than the count
Anchoring	A high top price raising what feels fair for the middle	Tversky and Kahneman (1974): random starting numbers of 10 and 65 produced median estimates of 25 and 45	Many Labs 1 (Klein and colleagues, 2014), using a later anchoring task: replicated across 36 samples; the original underestimated it. Li, Maniadis and Sedikides (2021), 53 willingness-to-pay and willingness-to-accept studies: random-effects correlation of 0.267, “smaller than the effects reported in early studies”, an upper bound	Robust for estimates, moderate for what people will pay
Centre-stage effect	Putting the target tier in the middle column	Valenzuela and Raghubir (2009): consumers believe options in the centre of a display are the most popular and choose them more often; weaker when people focus on their own preferences	One paper, abstract only; no independent replication opened	Free to apply, unproven at scale

Two points stand out. First, the effect most often sold as “pricing psychology”, the decoy, is the weakest. The most retold example is a subscription menu in which a print-only option priced the same as print plus web makes the bundle look like a bargain. Frederick and colleagues note that a best-selling book reports that decoy lifting the target from 32% to 84%, and that the published version by Kivetz and colleagues reported 43% to 72% with groups of 29 and 30 students; their retests found no attraction effect. Huber, Payne and Puto, who first described the effect in 1982, reply that it holds “when the conditions of the study are essentially replicated”. Both sides agree it has boundaries; the one that matters here is that real buyers experience real services rather than tidy matrices of numbers.

Second, the middle option’s pull has a catch. The extremeness-aversion meta-analysis reports that “employing price-quality tradeoffs” tends to reduce the effect, and a good-better-best menu is a price-quality trade-off by design. Expect a nudge toward the middle, not a default.

Which effects should a solo business design around?

• Clarity and the compromise position. Make the target tier the middle one and its advantages over the entry tier obvious. Simonson found the effects stronger “among subjects who expect to justify their decisions to others”, which is the normal position of a business buyer.
• A real top tier. In the camera study the expensive model was no decoy: it took 21% of choices. A premium tier people actually buy earns revenue and sets the reference point.
• Simplicity, not mere shortness. The overload meta-analyses point to complexity, not the number of options. Tiers that differ on one or two clear dimensions beat tiers with a dozen features each.
• No reliance on decoys. A tier that exists only to be rejected may do nothing, and Frederick and colleagues saw added inferior options sometimes cut the target’s share, a “repulsion effect”.

How do real platforms structure their tiers?

Annual reports filed with the US Securities and Exchange Commission carry legal liability and audited financial statements, so they are a firmer record than pricing pages, which change often and frequently block automated access.

Table 2. How platforms used by solo businesses structure tiers. CEOtudent analysis of SEC filings; figures as printed.

Company and filing	What is tiered	What the filing says	Lesson for a solo menu
Fiverr, 20-F for 2025	Seller gigs, sold as packages	Packages “tiered as Basic, Standard and Premium”, varying by word count, video length or revisions. Take rate of 27.7% of marketplace GMV, from commissions paid by buyers and sellers. Buyers who spent over $500 produced 66% of marketplace revenue	Tier on a countable deliverable. A minority of high spenders can carry most revenue
Shopify, 10-K for 2025	Merchant plans	Most merchants “subscribe to our Basic and Grow plans”, yet most GMV comes from Shopify Plus and enterprise merchants. Plus starts at “several times” the Advanced plan rate. Cohorts have “upgraded their subscription plans” over time	Entry tiers win on headcount, top tiers on value. Build the upgrade path on purpose
Wix, 20-F for 2025	Free plan plus premium packages	About 6.1 million premium subscriptions, about 83% on yearly or multi-year terms. AI credit limits follow “the package they purchase”. Online payments require a premium plan. Email marketing priced by emails sent per month	Fence one valued capability, meter usage by tier, reward commitment with term rather than scope
Upwork, 10-Ks for 2022, 2023 and 2025	Platform fees, by relationship size and client plan	2022: talent typically paid 20% of the first $500 billed to a client, 10% of the next $9,500 and 5% above $10,000, and maturing relationships pushed more volume to the lower rates. May 2023: a flat 10%. May 2025: 0% to 15% per new contract. Clients pay a 5% marketplace fee, or 10% on the Business Plus plan (less when paying by ACH)	Volume-discount tiers can erode yield as accounts grow. A premium plan can carry a higher fee when it adds features
Etsy, 10-K for 2025	Seller fees by sales volume	Offsite Ads fee of 12% or 15% of a sale “based on the seller’s volume of sales”, on top of a 6.5% transaction fee and a $0.20 listing fee	Tiers can follow customer size rather than features

Three patterns run through the filings. Tiers move along a unit buyers can count. Headcount and value diverge: Shopify’s volume comes mostly from its top plans, and Fiverr’s spenders above $500 produce two-thirds of its marketplace revenue. And structures get revised when the economics drift: Upwork’s report for 2023 says the switch to a flat fee, together with other fee initiatives, contributed to a $67.8 million, or 13%, rise in marketplace revenue. For a solo seller: re-run any discount ladder for your largest clients as they grow. For how platform fees compare with selling direct, see the freelance channel economics comparison.

How should you design the three tiers?

Fill in the “what is held back” row before the “what goes in” row: what a tier leaves out is what makes buyers move up.

Table 3. Tier design template. CEOtudent editorial framework.

Design element	Good (entry)	Better (target)	Best (premium)
Job of the tier	Lowers the barrier, qualifies price-sensitive buyers	Where you want most buyers; best value for delivery effort	Serves those who value speed, depth or access most; sets the reference point
Who it is for	First-time or budget-limited buyers	The typical buyer the offer was built for	Buyers with urgency, higher stakes or a budget holder who wants certainty
What goes in	The core outcome with one clear limit (one revision round, one deliverable, self-serve support)	Core outcome plus the two or three extras most buyers ask for anyway	Everything in Better plus a scarce input: faster turnaround, more access, wider scope or your direct involvement
What is held back	Revisions, speed, support, customisation	Your scarcest input, such as calendar priority	Nothing essential; limits are on volume so scope stays bounded
Pricing logic	Covers its own delivery cost and a margin	Contribution per delivery hour at or near your best	Priced on contribution per hour after realistic scope
Fence	A visible limit buyers can hit	The limit that matters most to the target buyer is removed	A capability only this tier has, as Wix ties payments to premium plans
Warning sign	Most buyers choose it	Fewer buyers choose it than choose Good	Nobody buys it, or it eats your calendar

Vary tiers along one or two dimensions, as Fiverr’s packages do. Name them by what the buyer gets (“Audit”, “Audit and plan”, “Audit, plan and launch support”) rather than by adjectives. And never list a tier you would be unhappy to sell.

What does the anchoring math look like?

A verified example first. In Simonson and Tversky’s camera test, the catalogue prices were $169.99, $239.99 and $469.99: the middle model cost 1.41 times the entry model, and the top model 1.96 times the middle. Had those hypothetical choices been purchases, the shift from 50/50 to 22/57/21 would have raised revenue per 100 choosers from $20,499 to $27,289, or 33.1% (CEOtudent calculation from the printed shares and prices). Two cautions: the choices were hypothetical, in a 1992 study, and part of the gain comes from the 21% who chose the top model.

The same paper reports a retail anecdote, not a controlled study: when Williams-Sonoma added a $429 bread maker beside its $275 model, the new item sold few units, but “the sales of the less expensive appliance almost doubled”. No study opened for this article identifies an optimal ratio between tier prices.

Then an illustrative solo menu. The inputs are ILLUSTRATIVE assumptions invented for this example, not market data. A consultant sells an audit in three tiers:

• Basic: $750, 6 delivery hours, $30 direct costs
• Standard: $1,800, 12 delivery hours, $60 direct costs
• Premium: $4,200, 24 delivery hours, $120 direct costs

Standard costs 2.4 times Basic, and Premium 2.33 times Standard. Contribution per delivery hour (price minus direct costs, divided by hours) is $120, $145 and $170. Capacity is capped at 100 delivery hours a month, with enough demand to fill it.

Table 4. Mix and contribution math under ILLUSTRATIVE assumptions. CEOtudent calculation; the mixes are assumed, not observed.

Illustrative mix (Basic/Standard/Premium)	Revenue per client	Delivery hours per client	Contribution per delivery hour	Monthly contribution, 100 hours	Same, if Premium takes 36 hours
Single offer, Standard only	$1,800	12.0	$145	$14,500	$14,500
A: entry-heavy, 50/40/10	$1,515	10.2	$144	$14,353	$12,842
B: target-centred, 20/60/20	$2,070	13.2	$152	$15,182	$12,846
C: top-heavy, 10/50/40	$2,655	16.2	$159	$15,889	$12,257

1. Revenue per client misleads a time-capped business. Mix C earns the most per client in both scenarios, but once Premium creeps from 24 to 36 hours its contribution per hour falls to $113 and the top-heavy mix produces the lowest monthly contribution of the four rows.
2. A menu beats a single offer only while the tier you push keeps its contribution per hour. B and C beat the single offer when scope holds; every tiered mix falls below it when Premium overruns.
3. An entry tier can be a hidden discount. Mix A earns slightly less than the single offer even before any overrun, because Basic pays least per hour; if those buyers would otherwise have bought Standard, the entry tier is a discount.

The CEO question is not “how high can the top price go” but “which tier should buyers choose, and does it pay per hour”. If Premium pays best per hour, protect its scope in writing; if not, reprice or narrow it.

Pre-launch checklist: is your menu ready?

Decision test. CEOtudent editorial framework. Launch when all five “must” items are a yes and at least three of the five “should” items are a yes.

Must:

1. Can you name the target tier, and is it the middle one?
2. Does each tier have a written price, delivery-hour estimate and direct cost, and a positive contribution on its own?
3. Does every tier differ from its neighbour on something a buyer can count or picture (revisions, turnaround days, sessions, seats, credits)?
4. Can you deliver Premium at its stated scope this month, for every buyer who picks it?
5. Is every price genuine and complete, including any “was” price, with no fees revealed only at checkout?

Should:

6. Does one valued capability appear only from the target tier upward?
7. Could a buyer explain in one sentence, to a manager or partner, why they chose the middle tier?
8. Can each tier’s scope be read as a short list in one look?
9. Have you decided what happens when a Basic buyer asks for Standard extras (an upgrade price, not a favour)?
10. Is tracking in place for tier views, choices and actual delivery hours?

Item 5 has a legal edge. The US Federal Trade Commission’s Guides Against Deceptive Pricing (16 CFR Part 233) treat a former-price comparison as legitimate when the former price is “the actual, bona fide price at which the article was offered to the public on a regular basis for a reasonably substantial period of time”. In the EU, Directive (EU) 2019/2161 inserted Article 6a into Directive 98/6/EC: an announced price reduction must show the prior price, “the lowest price applied by the trader during a period of time not shorter than 30 days” before it, with some room for national variation. Neither rule names decoy tiers, but a tier that is listed and cannot actually be bought raises the same problem: a reference price that is not real.

How do you test tier changes like a student?

A classroom result with groups of 29 and 30 students became the textbook decoy example, and a retest with 2,003 respondents did not find it. Your menu data will be smaller still, so discipline matters more than statistics. A five-step protocol (CEOtudent editorial framework):

1. Write the hypothesis and the decision rule first. For example: “Moving priority turnaround from Premium to Standard will raise Standard’s share without lowering contribution per hour.”
2. Change one element at a time: price, scope, fence, order or names.
3. Log every offer seen and every choice made, with delivery hours actually used and scope added after purchase. Shares without hours tell half the story.
4. Decide in advance how many buyers you will wait for. Small samples swing.
5. Keep the failures. The replication record in Table 1 exists because researchers published null results.

For digital products, run the test before launch by pre-selling tiers with the 7-day pre-sale protocol, and set price points with the digital product pricing architecture. There, watch support load and refunds per tier instead of delivery hours.

What this does not mean

• Three tiers do not always beat one offer. When capacity is tight or buyers are alike, a single offer can be the better business; see the one-product business.
• Lab results do not transfer directly. Most of Table 1 comes from consumer products and often hypothetical choices; services bought by businesses are far less studied.
• Platform structures are not optimal prices for a solo seller.
• The illustrative numbers are not benchmarks. Replace every input in Table 4 with your own.

Frequently asked questions

How many tiers should a solo business offer?
Three is common and easy to compare, but the overload evidence points to complexity, not the count. Two tiers can work; a fourth rarely adds clarity.

Which tier should be the target?
Usually the middle, because the compromise effect is the best-supported of the menu effects reviewed here. Make sure it also has your best or near-best contribution per delivery hour.

Is a decoy tier worth adding to push buyers up?
The evidence does not support relying on it: no significant attraction effect in 27 studies with experienced attributes, and 11 significant results in 91 attempts in another programme. Build a top tier people can actually buy.

How far apart should tier prices be?
No study opened for this article gives an optimal ratio. The 1992 camera steps were 1.41 and 1.96; the illustrative menu uses 2.4 and 2.33. Work from delivery cost and contribution per hour, then test.

Do tiered offers work for digital products?
Yes, through usage limits, credits and capabilities reserved for higher plans, as the filings show. The constraint shifts to support time and keeping tiers visibly different.

What should be tracked after launch?
Tier shares, contribution per delivery hour by tier, upgrade and downgrade requests, and scope added after purchase.

Can a premium tier show a “was” price?
Only if it reflects a genuine former price; the FTC guides and the EU prior-price rule both target reference prices never really charged. This is general information, not legal advice.

Sources

1. Huber, J., Payne, J. W., Puto, C. P. Let’s Be Honest About the Attraction Effect. Journal of Marketing Research 51(4), 2014.
2. Frederick, S., Lee, L., Baskin, E. The Limits of Attraction. Journal of Marketing Research 51(4), 2014.
3. Yang, S., Lynn, M. More Evidence Challenging the Robustness and Usefulness of the Attraction Effect. Journal of Marketing Research 51(4), 2014.
4. Simonson, I. Choice Based on Reasons: The Case of Attraction and Compromise Effects. Journal of Consumer Research 16(2), 1989.
5. Simonson, I., Tversky, A. Choice in Context: Tradeoff Contrast and Extremeness Aversion. Journal of Marketing Research 29(3), 1992.
6. Neumann, N., Böckenholt, U., Sinha, A. A meta-analysis of extremeness aversion. Journal of Consumer Psychology 26(2), 2016 (abstract only).
7. Iyengar, S. S., Lepper, M. R. When Choice is Demotivating: Can One Desire Too Much of a Good Thing? Journal of Personality and Social Psychology 79(6), 2000.
8. Scheibehenne, B., Greifeneder, R., Todd, P. M. Can There Ever Be Too Many Options? A Meta-Analytic Review of Choice Overload. Journal of Consumer Research 37(3), 2010.
9. Chernev, A., Böckenholt, U., Goodman, J. Choice overload: A conceptual review and meta-analysis. Journal of Consumer Psychology 25(2), 2015 (abstract only).
10. Tversky, A., Kahneman, D. Judgment under Uncertainty: Heuristics and Biases. Science 185(4157), 1974.
11. Klein, R. A., et al. Investigating Variation in Replicability: A “Many Labs” Replication Project. Social Psychology 45(3), 2014.
12. Li, L., Maniadis, Z., Sedikides, C. Anchoring in Economics: A Meta-Analysis of Studies on Willingness-To-Pay and Willingness-To-Accept. Journal of Behavioral and Experimental Economics 90, 2021.
13. Valenzuela, A., Raghubir, P. Position-based beliefs: The center-stage effect. Journal of Consumer Psychology 19(2), 2009 (abstract only).
14. Fiverr International Ltd. Form 20-F for fiscal year 2025. SEC filing, company-reported.
15. Shopify Inc. Form 10-K for fiscal year 2025. SEC filing, company-reported.
16. Wix.com Ltd. Form 20-F for fiscal year 2025. SEC filing, company-reported.
17. Upwork Inc. Forms 10-K for fiscal years 2022, 2023 and 2025. SEC filings, company-reported.
18. Etsy, Inc. Form 10-K for fiscal year 2025. SEC filing, company-reported.
19. US Federal Trade Commission. Guides Against Deceptive Pricing, 16 CFR Part 233.
20. Directive (EU) 2019/2161 of 27 November 2019, Article 2 (inserting Article 6a into Directive 98/6/EC). Official Journal of the European Union L 328.

CEOtudent analyses: Tables 1 and 2 are CEOtudent compilations of figures as printed in the sources; the camera price ratios, revenue per 100 choosers and the 33.1% change are CEOtudent calculations from Simonson and Tversky’s printed shares and prices; Table 3, the decision test and the testing protocol are CEOtudent editorial frameworks; Table 4 uses ILLUSTRATIVE assumptions, not data. The decoy figures of 32% to 84% and 43% to 72% are cited as reported by Frederick, Lee and Baskin; the original classroom sources were not opened. Deliberately not used: the Harvard Business Review article on good-better-best pricing (paywalled), any claim that a fixed tier-price ratio or “decoy pricing” lifts revenue by a stated percentage (no primary source found), and Upwork’s Project Catalog tiers, which the annual reports opened here do not describe.

---

This content was compiled with the support of AI following in-depth research, then written and prepared for publication by the CEOtudent editorial team.

This post is also available in: