The best business model for APIs
Mis à jour : janv 22
You cannot avoid speaking to customers
I have experienced two main reasons why we do not dare speaking to customers. The first one is that we are afraid to hear customers telling us giving us painful arguments and put us out of our comfort zone. The second reason is that management and especially Product Managers think they already have the answers. In both cases, this prevents you finding the right value and willingness to pay. However, this is a critical step.
Why value based pricing is the best business model
There are three main business models that can be used (cost based pricing, competitor-based pricing or value based pricing).
Cost based pricing (also called “cost plus pricing”) first requires to determine fixed and variable costs that are needed to build, maintain and deliver data offers/API products. The selling price is defined by adding a markup to the unit cost. This price does not represent at all the perceived value from your customers. That is why choosing this option would mean leaving a lot of money on the table.
Competitor-based pricing requires to analyze the price competitors set for similar data products. However, since data is unique , two offers are not really comparable. In addition, choosing this option would lead to a price war leading to a margin cut.
Value-based pricing means pricing data based on how much the target customers believe it is worth. It is the best solution to maximize your profit as it considers what value you offer and what each of your customers are willing to pay. As it is the best business model for data, it has different flavors.
How to implement value based pricing
Here are five different flavors or major options to apply value based pricing.
These five models are by no means the only ones you can pick. You can mix some of their attributes to create your own recipe.
Different flavors are a matter of taste
Market-based pricing is often the best way to determine the true market price at any time but is not appropriate for recurring revenue. Dynamic pricing is quite often used for booking websites (flights, hotels). In principle, it applies to any item that is limited and where the context objectively defines the value for one off transaction well. The approach works smoothly when demand and price elasticity vary significantly in your market.
However, data API requires regular touching points with data buyers. Alternative pricing is quite often the option data sellers choose for their APIs by measuring the number of API requests that are made within a timeframe. Therefore, they average out the potential value they provide to a customer segment dividing it by the number of calls. It means that a buyer would pay the same price for each data request he makes whatever the value provided. Therefore, risk is then bore by API buyers who need to have enough value from the API (expected and perceived value) compared to the incurred costs. Therefore, there is a perceived risk from the buyer as he might not get enough value compared to the costs (generated). Buyers need to average out the value and take the risk in mind when buying an API.
Consequently, the cost per call is lower than it should be in order to compensate this risk for the buyer. Again, money is left on the table… More companies would access if better pricing/packaging options if only they existed if the balance between costs and value was tailored to their needs.
However, remember that the data value is contextual and APIs require ongoing relationship. The best blend of monetization should value the attributes of data and APIs:
Recurring relations by sending regularly data payloads
Data is unique
Value of data is contextual (same data result can have different value)
An interesting way to look at this is to map out those monetization models based on two axes:
Capture most value per transaction
Fitting an ongoing relationship
Therefore, the best approach for data is to have a subscription model with alternative/dynamic pricing options to capture most value. This alternative/dynamic pricing would help to define different price points within each segment.
The “Data value based pricing” approach helps sellers to capture a fair share of the perceived value. It reduces the risk as well for data buyers as they pay for the value they get. Data buyers would pay for performance rather than data requests. They would have the flexibility to manage costs appropriately in good and bad times.
This business model implementation allows to charge more to customers who are willing to pay for higher-end version of API products—and less to those who are looking for something simpler. This increases the market share and maximizes the revenue per customer. BCG (Boston Consulting Group) states that having this granular packaging and pricing to fit the needs of customers can increase API products returns over 25%.
To create different API products by bundling together several data points adequately, you must follow some rules so that low priced API products do not cannibalize high priced ones. API products must be communicated efficiently to customers. In the next blog post, we will review that in detail considering pitfalls and we will also focus on the best method to achieve it.