Improving Observability Transparency and Predictability with Consumption Pricing and Billing

By Mark Crawford, VP, Strategy and Execution, New Relic.

  • Posted in

Considering some offices were still using typewriters thirty years ago, businesses’ software stacks have come a long way in a short amount of time. Now, software is not just intrinsic to the proper functioning of an organisation, it is necessary for them to function at all. It underpins online presence, customer experience and productivity. Without properly functioning software most businesses would simply cease to operate.

A comprehensive understanding of how software stacks perform in real-time is critical, which is where observability enters the picture. It has become an important practice for modern digital enterprises, enabling engineers to catch errors before customers experience them, ensuring the brand can deliver a seamless experience. New Relic’s 2022 Observability Forecast revealed that over 50% of IT decision makers see observability as a key enabler for hitting core business goals.

And currently, cost saving for business is as important as it has ever been. The economic climate has hit businesses hard, impacting disposable incomes and revenue. When it comes to observability, finding opportunities to maximise cost-efficiency, while still providing themselves with the necessary visibility, has become a priority for many businesses.

Paying over the odds

Traditional pricing models for observability were rigid. Host-based pricing, or infrastructure-based as it is also known, is quantity based. Observability platforms that use this model bundle together product pricing for each individual type of monitoring on a monthly use-it-or-lose it basis, spreading costs across multiple pricing metrics, such as hosts, nodes, agents, or containers.

Modern, cloud-managed architectures often have abstracted underlying infrastructures. This makes it difficult to count hosts, thereby making it harder to predict costs. Many observability platforms also require customers to bundle adjacent use cases, another way that they will end up having to pay for extra monitoring they may not necessarily need.

The overarching result is that businesses that use host-based pricing for observability often see prices driven up outside of their control. Due to this, engineering teams have to reduce their instrumentation coverage and the amount of telemetry data they ingest. They would have to sample data, count agents and create custom metrics. This would all happen while worrying that they might be hit with surprise or hidden fees or instead, having to curb infrastructure growth so that it stayed in line with observability budget requirements.

In short, previous pricing models didn’t match the way technology operates today. Data usage is flexible and businesses can scale up overnight, while worldwide external factors – such as the pandemic – can cause data usage to drop just as quickly. Suitable pricing models should also be able to flex with the market in which a business finds itself. In response to this, New Relic changed its pricing and billing model to a consumption-based one.

A new pricing model

The consumption model has been welcomed by businesses at a time when every penny spent must be justified. By reducing the pricing metrics to just two, the number of users and the amount of data used, businesses are able to pay for only what they use, eliminating shelfware. Customers can also benefit from free usage credits (freemium) that allow them to try before they buy without any risk. A 2021 playbook by OpenView found that a usage-based model allows customers to gradually raise their expenditure and directly link the price paid with the value received.

This model also helps to strengthen channel relationships. Vendors are incentivised to remain invested in their customers’ success, as it is to their benefit to ensure customers derive value from the solution, increasing consumption.

According to New Relic’s 2022 Observability Forecast, flexibility to scale usage on consumption with no monthly minimum is the most important billing feature for customers. Indeed, many new customers – including Tibber, a European energy-based company, and Tesco, one of the UK’s largest retailers – have cited New Relic’s consumption pricing model as a reason for signing on with the observability company.

The need for observability

Increased adoption of continuous deployment practices and microservice-oriented architectures make observability critical to modern businesses. Proper use can help organisations discover and address ‘unknown unknowns’, monitor issues across the entirety of the software lifecycle, and work in conjunction with AIOps to create automated radiation of problems.

The 2021 Observability Forecast found that 65% of respondents were shifting away from legacy subscription-based models, such as host-based pricing, to a consumption model (23% of respondents). This fundamentally changes the relationship between the vendor and the customer – creating longer-lasting relationships – while dramatically improving the value for the customer.

Consumption-based models allow businesses to expand their observability practices as needed, while helping them to monitor and control spending. Observability can act as a silent guardian, keeping a continual eye on operations – as long as its scope isn’t limited. By switching to consumption-based pricing, businesses can more easily monitor the entirety of their software stacks.

By Dr. Atif Farid Mohammad – Head - AI R&D Center of Excellence, Apexon.
Procurement expert Keith Hausmann highlights how NLP and machine learning are driving efficiencies and savings while enabling sourcing teams to add more business value.
By Peter Greiff, Data Architect Leader EMEA, DataStax.
Keerthi Mukku, Commercial Account Manager at Plextek investigates the importance of Foreign Object Detection (FOD) at any airport.
Josip Rozman, AI consultant at Plextek examines if we should be using AI to write our content.
By Rohit Bhosale, Persistent Systems.
By Kirsty Biddiscombe, UK Head for AI, ML & Analytics at NetApp.
By Mark Lowe, business development director at North.