Pricing Architecture — Tiers, Metrics, and Anchors
Pricing is not a number — it is an architecture that shapes how customers buy, how you upsell, and what the market believes about your positioning.
🎯 By the end you'll understand
What a value metric is and why choosing the wrong one will cap your revenue ceiling permanently. How three-tier packaging works psychologically and structurally, what each tier is actually doing, and where the "magic middle" comes from. How anchoring and decoy pricing shape willingness-to-pay in B2B deals. And the specific failure mode that kills mid-market SaaS pricing: a tier architecture that cannibalizes itself because the entry tier delivers 90% of the value for 30% of the price.
KEY INSIGHTPricing architecture is the skeleton your revenue model grows around
Most technical founders treat pricing as a number they pick at the end. The damage from this mistake persists for years: the wrong value metric locks you into a ceiling, and a self-cannibalizing tier structure trains the market to buy down instead of up.
You already know from Lesson 3.2 that price should be set from economic value, not cost. Now the question is the architecture around that price: what unit does the customer buy, how do you package it into tiers, and how do you structure the choice so the most valuable option looks like the obvious one? These are three separate decisions, and each one compounds or destroys the others.
Every pricing architecture answers three questions in sequence. First: what is the value metric — the unit of value the customer buys, which should scale naturally with their size and usage? Second: how do you package tiers — what features live in each tier, at what price, for which buyer? Third: how do you anchor — what reference points shape the customer's sense of whether the price is high or low? Answer them in the wrong order, or answer any one badly, and the whole structure underperforms.
CORE CONCEPTThe value metric — what you actually sell
A value metric is the unit that scales with the value the customer gets. Charge per seat if value scales with how many people use the tool. Charge per site if value scales with physical locations. Charge per API call if value scales with transactions. Charge per dollar of spend managed if value scales with the size of the portfolio you're optimizing.
The critical test: does your value metric naturally expand as the customer grows? A customer who starts small should pay less than a customer who is large — and the growth between them should be revenue for you, not a renegotiation. If your value metric passes this test, revenue growth is baked into the product. If it fails, every expansion requires a sales conversation and a budget fight.
| Value metric | Natural expansion driver | Best fit when… | Risk |
|---|---|---|---|
| Per seat | Team grows, adoption spreads | Collaboration tools, productivity | Customers cap seat count; share accounts |
| Per site / location | Physical footprint grows | Physical-world software (retail, mfg, logistics) | Slow expansion; land-and-wait dynamic |
| % of value delivered | Customer outcomes improve | Cost savings or revenue optimization tools | Hard to measure / dispute causation |
| Flat platform fee + consumption | Volume increases | Infrastructure, data platforms | Customers optimize to reduce consumption |
| Annual platform + per site | Both software adoption and physical growth | Hardware+SaaS hybrid (Forge model) | Complex to explain; two line items in PO |
The wrong value metric is usually irreversible without a re-pricing event that chases away customers. Choosing it badly early is one of the five ways technical founders leave real revenue permanently on the table. The right answer usually emerges from understanding which dimension of customer size most strongly predicts the economic value they get from your product — and that comes from willingness-to-pay research (Lesson 3.4), not instinct.
Ask yourself: if a customer doubles in size (more plants, more transactions, more users, more revenue), does your value metric automatically double their contract? If yes, you have a good value metric. If no — if the customer can double in the dimension that matters while paying you the same flat fee — you have a ceiling problem. Fix the value metric before you build the tier architecture around it.
CORE CONCEPTThree-tier packaging — what each tier is actually doing
Three-tier pricing is not about offering different amounts of the same thing at different prices. Each tier serves a structurally different purpose. Understanding the purpose of each tier tells you what belongs in it, who buys it, and what "success" looks like for that tier — which is rarely just "sell more of it."
Entry tier — defines the floor and qualifies buyers
The entry tier exists to lower the activation energy for a purchase. Its job is to reduce friction, not to generate maximum revenue. In B2B, an entry tier that is too cheap attracts non-buyers — organizations that will never expand because they don't have the complexity to justify more. An entry tier that is too expensive kills the top of your funnel. The structural requirement: the entry tier must deliver real, measurable value to the buyer segment it targets. If it doesn't, you attract churners. If it delivers too much value relative to the mid tier, you cannibalize yourself — the critical failure mode this lesson addresses directly.
Mid tier — the revenue center and the product anchor
The mid tier is where most B2B SaaS revenue actually lives. It is priced to capture the most common buyer profile — the customer who has the problem seriously enough to pay meaningfully, but is not yet at enterprise complexity. The mid tier is also the psychological anchor: when a buyer sees three options, their eye goes to the middle first. They evaluate both ends relative to it. This means mid-tier pricing anchors the perception of value for the whole table. Under-price the mid tier and you depreciate the enterprise tier. Over-price the mid tier and you push the volume of deals into the entry tier, where they may never expand.
Enterprise tier — the ceiling, the signal, and the negotiation anchor
The enterprise tier is rarely the majority of deal volume — especially early. Its job is threefold. First, it is a real product tier: customers who need deep integrations, dedicated SLAs, custom reporting, and compliance documentation genuinely belong here and pay for it. Second, it signals market positioning: a $150K/year enterprise tier communicates to the entire market that this is not a $12K/year point solution. Third, it is an anchoring device — the enterprise tier price makes the mid tier look reasonable. A $42K/year deal feels different when it's next to a $120K/year option than when it's the most expensive thing on the page.
Here is the specific failure mode this lesson is built around: a tier architecture where the entry tier delivers 85–90% of the value for 25–35% of the price of the mid tier. When this happens, rational buyers — especially price-sensitive ones — choose the entry tier and never upgrade. They don't need to. The upgrade path looks like paying 3x more for 10% more functionality. This is not a sales problem or a customer success problem. It is a packaging architecture problem, and the fix requires redesigning which features belong in which tier — not a discount or a better pitch.
HOW IT WORKSAnchoring and decoy pricing in B2B
In B2B, buyers have weak reference prices for software. A manufacturing ops manager has no intuition about whether $42K/year for an AI quality inspection platform is expensive or cheap — unlike a consumer buying a phone, they have no market-clearing price in their head. This makes B2B pricing unusually susceptible to anchoring: the first number the buyer sees shapes all subsequent perception of value.
Three anchoring mechanisms matter in B2B pricing architecture:
The high anchor — what makes mid look reasonable
When a buyer sees a $120K/year enterprise option before seeing your $42K/year Growth tier, the $42K feels modest. When there is no enterprise tier — when $42K is the most expensive option — it feels expensive. This is pure anchoring psychology and it is fully documented in the behavioral economics literature. The strategic implication: even if you expect to close zero enterprise deals in year one, the enterprise tier price on your pricing page is doing real work. It is making your target price feel accessible by comparison.
The decoy effect — a third option that makes one of the others dominant
The classic decoy effect (Ariely and others have documented this across many consumer contexts) involves a third option that is dominated by one alternative but not the other. In B2B pricing, this often looks like: Starter at $15K, Growth at $42K, Enterprise at $45K — where Enterprise has marginally more than Growth but almost the same price. Virtually every buyer chooses Growth because it looks nearly as good as Enterprise for $3K less. The Enterprise option is a decoy that makes Growth dominant. This is a legitimate design technique, but it requires the tiers to be priced deliberately — not just set based on cost.
Cost comparison framing — anchoring against the alternative
In value-based B2B pricing, the most powerful anchor is not another tier — it is the cost of the problem you solve. If a single undetected defect escape costs a manufacturer $140K, and you charge $42K/year to prevent them, the relevant reference is $140K, not your competitor's $38K/year price. Presenting the value anchor first — before presenting your price — reframes the entire budget conversation. Sales reps who skip this step are leaving the buyer anchored to their previous software costs, not to the value being replaced.
For a three-tier architecture to hold without constant discounting, each tier needs to deliver a perceived value ratio of at least 1.5–2x the price step. If Growth costs 3x Starter, it needs to feel like it delivers 4–6x the value, not 3x. If it only delivers 2x the value at 3x the price, buyers will consistently choose Starter and blame "expensive pricing" for what is actually a packaging problem. The ratio test: write down the three most important features in each tier and ask which buyer segment genuinely needs them. If every buyer needs all the Growth features to get value, Starter should not exist. If no buyer needs the Growth features beyond what Starter offers, your tiers have not been correctly partitioned.
SEE ITPricing Tier Builder — construct Forge's three-tier architecture
Notice what happens when you over-populate Starter with high-value features. The upgrade rate collapses — a customer with real-time defect detection AND basic predictive alerts in Starter has no reason to move to Growth just to get "custom reporting." That's the cannibalization signal. The correct fix is not to raise the Starter price — it's to move features up. Specifically: the features that correlate with customer size and usage growth belong in Growth. Features that any customer needs on day one belong in Starter. The line between them is the value metric made architectural.
FORGEPackaging Forge's pricing architecture
By this point in Stage 3, Forge has converted two pilots to annual contracts at $42K ACV each — $30K platform fee plus $12K hardware (one-time per site). That is the Growth tier deal. Now the question is: what do the Starter and Enterprise tiers look like, and do they hold together as a coherent architecture?
Forge's value metric: annual platform fee + per-site expansion
The platform fee ($30K/year) covers the SaaS dashboard, alerting, and analytics. The hardware fee ($12K one-time per site) covers the edge sensor deployment. This is a two-component value metric that scales naturally: a customer who adds sites generates more hardware revenue plus a potential platform tier upgrade. The risk is complexity — two line items in a PO require two budget approvals. The mitigation: bundle the first-year hardware into an all-in annual deal for new customers, preserving the expansion architecture without the friction of a two-step procurement.
Forge Starter: single-site, ops manager, $18K/year
Starter covers one production line, core defect detection, a basic dashboard, and email alerting. Hardware ($12K) is included in year one as a bundle. The buyer is a plant ops manager or quality engineer at a facility that runs 1–2 lines and wants to run a pilot without a procurement committee. Critical packaging decision: Starter does NOT include predictive maintenance alerts or custom reporting — these require data from multiple lines over time, which naturally belongs to Growth customers. This prevents the cannibalization failure mode: a Starter customer genuinely needs Growth's multi-site capabilities to access predictive value.
Forge Growth: 3–8 sites, ops director, $42K/year platform + $12K/site hardware
This is the Forge milestone deal. Growth includes multi-site dashboards, predictive maintenance models, root-cause analysis, API access for ERP integration, and quarterly business reviews. The buyer is a Director or VP of Operations with multi-site responsibility. The upgrade path from Starter is clear: once a customer has validated value at one site and wants to roll out to additional lines, they need Growth's multi-site dashboard and ERP integration. The price step ($18K → $42K, a 2.3x increase) needs to feel like a 4x+ value step — which it is, because predictive maintenance and ERP integration are worth far more than basic defect detection alone.
Forge Enterprise: 10+ sites, C-suite, $120K/year platform + hardware
Enterprise adds dedicated SLA (99.9% uptime guarantee with contractual remedies), custom compliance reporting for ISO/IATF automotive standards, dedicated customer success manager, multi-tenant security, and custom model training on the customer's own defect library. The buyer is a COO or VP of Global Operations at a tier-1 supplier with 10+ manufacturing sites globally. Even if Forge closes zero Enterprise deals in Year 1, this tier is doing critical work: it signals to the $42K buyer that this is enterprise-grade software priced accessibly, not a startup product priced optimistically.
The critical question at each tier boundary: does a customer in the lower tier genuinely need to grow (in size, complexity, or ambition) before they need the upper tier? For Forge: a single-site customer truly cannot use multi-site dashboards. They need to add sites to unlock predictive maintenance value across lines. That is a natural upgrade trigger, not an artificial feature withholding. Contrast this with the failure mode: if basic predictive maintenance were in Starter, a single-site customer could get substantial maintenance value without ever adding a second site. The tier boundary would collapse.
WEAKNESSWhere this bites
The most common B2B pricing architecture failure: a Starter tier that delivers 85–90% of the product's core value. Technical founders build this because they want to be generous, or because they genuinely believe their core feature is table stakes. The result is an expansion problem masquerading as a churn problem. Customers don't churn from Starter — they stay forever, happily, because they have everything they need. The ARR per customer never grows. The expansion playbook fails. The fix: identify the features that scale with customer size (multi-site dashboards, ERP integration, compliance reporting) and make those Growth-tier features from day one, even if they are technically easy to build into Starter. Packaging decisions are not about engineering complexity — they are about which buyer has the organizational need for which capability.
Some founders build an Enterprise tier with features they haven't built and deals they don't expect to close — purely as an anchoring device. This is a mistake for two reasons. First, if any buyer actually wants Enterprise, you have committed to delivering capabilities you cannot deliver, potentially on a timeline that destroys the relationship. Second, sophisticated buyers in manufacturing and enterprise SaaS will sometimes ask for an Enterprise reference customer. If you have none, the tier feels like theater, which it was. Build Enterprise tier capabilities you can actually deliver, for buyers you can actually identify. The anchoring still works — but it works with real proof points behind it.
A three-tier pricing page built for inbound self-serve does not work in an outbound enterprise sales motion — and vice versa. In enterprise B2B manufacturing, buyers do not visit a pricing page, pick a tier, and buy. They request a demo, receive a custom proposal, and negotiate terms. If your pricing architecture is not designed for the sales motion your buyers actually use, the tiers become irrelevant — every deal starts at zero and gets constructed from scratch, which means your anchoring and packaging never fire. The design of pricing architecture and the design of the sales process are joint decisions. Stage 4 covers the sales motion; Stage 3 should be building a pricing architecture that hands off cleanly to that motion.
KEY INSIGHTHow this connects
- 3.1 · Business Model Archetypes — your business model archetype determines which value metrics are even available to you. A hardware+SaaS hybrid like Forge cannot use a pure per-seat metric — the physical dimension of the product requires a per-site or per-device component. The archetype constraints you locked in 3.1 are upstream of every packaging decision made here.
- 3.2 · Value-Based Pricing — the prices in your tier architecture must be grounded in the economic value you deliver, not cost markup. The Growth tier at $42K only makes sense if the buyer's alternative (manual inspection labor + quality escapes + rework cost) is measurably more expensive. If you didn't do the value-based pricing work from 3.2, your tier prices are guesses that will collapse under the first serious negotiation.
- 3.4 · Willingness-to-Pay Research → — the tier structure you designed here generates hypotheses; willingness-to-pay research tests them. You have not validated that buyers will pay $42K for Growth features until you run structured pricing conversations. Lesson 3.4 gives you the methodology to measure what customers will actually pay before you lock in the architecture.
Pricing architecture is not set-and-forget. The tier structure you design at $42K ACV may need restructuring at $150K ACV as you understand your buyer's upgrade triggers better. The principle that persists across every revision: features belong in the tier of the buyer who naturally needs them, at the point in their journey when their organizational complexity actually requires them.
GLOSSARYKey terms introduced
- Value metric
- The unit of value the customer buys, which should scale naturally with their size, usage, or outcomes. A good value metric generates expansion revenue automatically as the customer grows, without requiring a renegotiation. Examples: per seat, per site, per API call, percentage of managed spend.
- Tier cannibalization
- The failure mode where the entry tier delivers so much of the product's core value that customers in the mid-market have no functional reason to upgrade. Results in permanent suppression of ACV and expansion revenue, often misidentified as a churn or sales problem when it is actually a packaging architecture problem.
- Anchoring (pricing)
- The use of a reference price to shape a buyer's perception of whether a target price is high or low. In B2B, the enterprise tier anchors the mid-tier price; the cost of the problem being solved anchors the entire package. Anchoring is most powerful in markets where buyers lack established reference prices.
- Decoy effect
- A pricing structure where a third option is introduced specifically to make one of the other options look dominant (clearly superior in value/price ratio). The decoy is not intended to sell — it is intended to guide the buyer toward the target option by comparison. Must be used carefully in B2B; if the decoy is discovered to be a design tactic, it damages trust.
- Perceived value ratio
- The ratio of perceived value to price for a given tier. For a tier architecture to hold without heavy discounting, each price step needs to feel like it delivers more value than the price multiple would suggest. A 3x price step needs to feel like at least 4–5x the value step, or buyers will consistently choose down.
- Expansion revenue
- Revenue generated from an existing customer beyond their initial contract — either through tier upgrades, additional sites, more seats, or higher usage. Expansion revenue has no CAC and is the primary driver of NRR (Net Revenue Retention) above 100%. A good value metric and tier architecture make expansion revenue structurally inevitable rather than dependent on sales intervention.
- Feature partition
- The deliberate assignment of product features to tiers based on which buyer segment genuinely needs each feature at what stage of their organizational growth. Correct partitioning means entry-tier customers naturally outgrow their tier as their complexity grows — not because features are artificially withheld, but because the upper-tier features require organizational scale to use effectively.
ACTIVE RECALLCheck your understanding
Answer in your own words first, then reveal.
1Forge charges $30K/year for the SaaS platform plus $12K one-time per site. Is this a good value metric? What is the natural expansion trigger, and what is the main structural risk?›
It is a reasonable value metric for a hardware+SaaS hybrid, but it has a specific structural risk. The natural expansion trigger is new site deployments — each time a customer adds a production line, they pay $12K hardware plus potentially trigger a platform tier upgrade. This is good because it creates automatic expansion revenue tied to customer growth. The structural risk is procurement complexity: two line items ($30K SaaS renewal + $12K hardware per new site) require two separate budget approvals — one from the IT/software budget and one from the capex or facilities budget. This creates friction in the expansion motion. The mitigation is bundling: quote the all-in year-one cost as a single number ($42K for one site), making the components visible in the contract but presenting a single number for the approval process. Expansion deals (adding sites) then quote as a single incremental number rather than forcing a new two-item procurement exercise.
2A founder notices that 70% of their Growth-tier deals are being closed at Starter price after the buyer objects to the Growth price. They conclude their Growth tier is too expensive. What is the more likely diagnosis, and what is the correct fix?›
The more likely diagnosis is not that Growth is over-priced — it is that Starter is over-featured. If 70% of Growth-caliber buyers are choosing Starter after seeing the price, it means the Starter tier contains enough value that buyers can justify the downgrade. They are making a rational decision: "I can get 80% of what I need for 40% of the price." That is not a pricing problem — it is a feature partition problem. The correct fix is to audit which features in Starter are being used by multi-site, expansion-intent customers and move those features to Growth. Specifically: any feature that requires multi-line data, organizational coordination, or integration with systems a single-site plant would not have should be a Growth feature. Lowering the Growth price would solve the symptom but leave the root cause intact: Starter would still cannibalize Growth because the feature set is wrong, not the price.
3Why does the Enterprise tier have pricing value even if you close zero Enterprise deals in year one?›
The Enterprise tier does three things for deals that close at Growth tier prices. First, it serves as a high anchor: when a buyer sees $120K/year before they see $42K/year, the $42K feels accessible rather than expensive. Remove the Enterprise tier and $42K becomes the ceiling, which changes the psychological framing entirely. Second, it signals market positioning: having a $120K/year tier tells every buyer that this product serves enterprise customers with genuine enterprise complexity — it is not a startup tool being sold up-market. Third, it provides a future upgrade path: a Growth customer who grows to 10+ sites now has a named, priced destination to grow into, which makes the enterprise conversation less surprising when it happens. The Enterprise tier earns its place on the pricing page through anchoring and positioning before a single Enterprise deal closes.
4In B2B pricing, why is "anchoring against the cost of the problem" often more powerful than anchoring against a competitor's price?›
Competitor anchoring sets up a relative value comparison ("we're better than X at the same price") — which is inherently a zero-sum contest where the buyer is still asking "is this worth the price?" Problem-cost anchoring reframes the question entirely: "is the cost of not solving this worth more than our price?" If a single quality escape event costs a manufacturer $140K in recalls, rework, and customer penalties, then $42K/year is evaluated against $140K, not against a competitor's $38K/year. The buyer's mental math becomes: "if this prevents even one escape event per year, the ROI is 3x." Competitor anchoring cannot achieve this reframing because it keeps the buyer's attention on the price paid rather than the cost avoided. The prerequisite for this technique — and why Lesson 3.2 must come before Lesson 3.3 — is knowing the actual economic cost of the problem, which requires genuine willingness-to-pay and value-based pricing research with real buyers.