Linear | Partial Tariff | Tiers, Bundling, & Discounts
[9 min. read]
How you structure pricing can impact value, perceptions, and WTP.
According to Madhavan Ramanujam, “How you charge is often more important than what you charge.”
Pricing structure helps create a match between both buyer and seller as they each seek to optimize the sale for themselves.
Buyers optimize for WTP, the quantity they need, and structures that match their perceptions of value. Sellers use structure to maximize average revenue per user (ARPU), a transactional metric for measuring the value-maximization strategy from Chapter 3.
5a. Linear (One-Part) Structure
That’s not to say that pricing structures are always fair or equal for both sides, just that they exist as a function of the interests of both.
The simplest structure is linear.
When something increases proportionately with each additional unit, it’s called marginal. Since linear pricing creates a one-to-one relationship between quantity and price, the entire structure is marginal.
In other words, if a grocery store sells apples for $1 each, customers pay one dollar for every apple they buy. The price per unit never gets greater and never gets less.
Gasoline prices are also linear. The price itself changes often and the price is different for regular versus premium, but the structure is always linear, a set price per gallon.
Linear structure sometimes doesn’t capture all of the advantages that best meet the needs of either buyer or seller.
Beyond linear there are multi-part structures, tiers, discounts, and bundling which offer more options for matching customer WTP.
5b. Partial Tariffs
In a two-part tariff system, customers pay a fixed price, plus a marginal price for each unit.
Since the fixed price is required in order to get the product units, it’s viewed as a tax or a tariff on the transaction.
Using apples again, if the supermarket required each customer to purchase a basket for five dollars and then to pay $1 for each apple they put in the basket, you’d have a two-part structure.
Similarly if a customer pays a flat fee to access a software platform, plus an additional fee per user, per transaction, or per process? That’s also a two-part structure.
Three-part pricing requires the same flat fee tariff to unlock the purchase of marginal units but the flat fee includes some units.
Instead of buying just a basket for $5, your basket purchase might include 5 “free” apples for $8. Then each additional apple costs $1. That’s a three-part tariff structure. The three parts being the basket, the included allotment of apples, and the additional apples purchased.
In SaaS software, three-part pricing has proven to generate the highest revenues. See the research supporting this in Yong Chao’s Strategic Effects of Three-Part Tariffs Under Oligopoly.
One reason customers prefer this is called the Taxi-Meter Effect. Research suggests that passengers in taxis experience anxiety from watching the meter tick up during their journey. By contrast the certainty of a fixed price is comforting to them. Purchasing in larger increments also reduces the cognitive load of making a purchase decision for every additional unit.
5c. Tiers, Discounts, and Bundling
The three structures are really just collections of tiers and discounts.
These constituent parts can be as simple or as complex as needed. (The tariff component is basically a minimum tier.)
In the GoReact example from Chapter 4, there were at least four clear bands of WTP—a great data-informed place to start evaluating pricing tiers.
Tiers may line up with WTP, which can map to certain sets of features. Other times tiers are tied to discounts.
Discounts for products with high elasticity are one of the most common and visible pricing policies.
Discounts can be permanent parts of a pricing program. For example discounts for volume, contract length, future commitments, or early payment may be standard practice.
Other discounts are temporary, seasonal, or promotional. Black Friday sales, Amazon Prime Day, back-to-school sales, holiday sales, product launches, and end-of-quarter incentives all fit this description.
And discounts come in many forms: percent off, dollars off, buy-one-get-one, buy-one-get-discount, free shipping, loyalty rewards and cash back, etc.
Consider a purchase of printed materials. Most printers charge a set-up fee and then a price per unit. Below is a pricing table from Moo’s website. Notice how the price per unit drops from $0.78 for fifty flyers to $0.19 for one thousand?
Where might this fit in the examples we’ve given?
It’s not linear for sure. It has some attributes of two-part tariff pricing, but there’s no flat fee. This is discounting based on tiers.
It’s also likely cost-plus because printing is fairly commoditized. Since it takes some work to set up a print run, but after that, the cost to create one more unit is proportionately tiny. Chances are the printer even runs a few extra for free and uses them as labels on the outside of the box.
Another way to use tiers is bundling.
Bundling like items is about volume (ever try buying just one egg?) Both the two- and three-part tariffs are a type of bundling.
Pizza and Breadsticks
You can also bundle complementary items like pizza and breadsticks. In the apples example, the basket is complementary to the apples. Likewise for software the software platform fee is complementary to the users or transactions made on the platform.
Madhavan Ramanujam has a pricing puzzle I love that uses pizza and breadsticks. I’ve often used it in my classes for MBA and MSF students at the University of Utah. Ramunajam reports that fewer than 10% of executives arrive at the most profitable approach.
Here’s the punchline. In Ramanujam’s scenario, adjusting the bundling strategy has the potential to unlock 54% more revenue.
Same product. Same mix of WTP. Same number of potential customers. 54% more revenue!
The only difference is the pricing approach.
Does that pique your interest?
Just think what you would be willing to do in order to raise revenue by 54 percent?
What would you do to raise ANY key metric by 54 percent?
That is what makes the headache of trying to uncover your customer’s true WTP totally worth it.
Okay, what is the pizza and breadsticks scenario?
Imagine you own a world famous pizzeria that makes delicious pizza and cheese breadsticks. In an effort to decipher pricing, you study your customers and discover four segments.
Segment A loves pizza and will pay as much as $9 for it but is not a big fan of breadsticks (they’ll only pay $1.50 for this).
Segment D loves breadsticks and will pay $9 but doesn’t love pizza (they will only pay $2.50).
The other two segments, B and C, fall in between A and D in WTP.
There are 100 of each segment coming into your pizzeria each month.
For example, if you sell the pizza at $4.50 and the breadsticks at $5, you would make a total of $2,850, since at $4.50, segments A, B, and C will buy the pizza; at $5 for the breadsticks, segments B, C, and D would buy it. Total revenue would equal ($4.50 × 300) + ($5 × 300) = $2,850.
What is the maximum monthly revenue you can earn?
The most common answer I get from MBA students is to charge $8 for pizza and $8.50 for breadsticks. They quickly see that maximizing revenue is a balancing act between number of customers and revenue per customer.
This approach yields ($8.00 × 200) + ($8.50 × 200) = $3,300.
A nice bump from $2,850.
But what if we priced based on combined WTP and bundled pizza and breadsticks together and sold the bundle for $10.50? This price is below the combined WTP for ALL customers.
It yields ($10.50 × 400) = $4,200.
In this example, bundling clearly generates more revenue than an unbundled price.
But there’s one more approach called mixed bundling. What if we offer either a bundle OR individual items from the menu? Could we capture more of the WTP?
If we bundle for $13 we capture the combined WTP for Segments B and C. Then if we price pizza or breadsticks alone at $9, we capture the WTP for pizza from Segment A and breadsticks from Segment D.
And the approach yields ($9 × 200) + ($13 × 200) = $4,400.
In this fictional scenario we went from $2,850 to $4,400 (a 54% increase) just by changing our pricing decisions.
Leader, Filler, Killer
What if the items you’re bundling are more diverse?
In the pizza and breadsticks example, both items are treated as equivalent. But what if we’re talking about a burger and fries? Would anyone place the same value on the fries as the burger?
The Leader, Filler, Killer framework (from Simon-Kutcher Partners) is this.
Leaders are must haves. These are the primary drivers of WTP.
Fillers are nice-to-haves. Customers like them, but probably wouldn’t pay more for them.
Killers are detractors, actually reducing WTP.
The example Simon Kutcher gives is a burger meal combo. The burger is the Leader, the fries and drink are Fillers. If the meal included coffee or a dessert, that would be a killer because many people don’t drink coffee or don’t want a dessert with their lunch.
Kevin Cohn adds that using this framework for “good, better, best” pricing for SaaS products (where each upgrade includes everything below it) means that a feature’s status as leader, filler, or killer changes depending on the bundle.
What is a leader for the “good” package, is by definition filler for the “better” package (otherwise you could charge “better” package prices for “good” features). Likewise, the leader feature in the “best” bundle is a killer for the “good” and “better” bundles because it’s more than they need or are willing to pay for.
One last caution about bundling. Consumers tend to average the prices in a bundle. So it’s usually not a good idea to bundle a large ticket item with a small ticket one.
To arrive at the best structure, tiers and/or bundling just examine the approaches in this chapter for your own company and product(s).
Is linear, two-part tariff, or three-part tariff the right structure? What impact on ARPU would there be by introducing pricing tiers and volume or other discounts? Finally, is maximum WTP captured with unbundled, bundled, or mixed bundle pricing?