October 21, 2019
When we announced Cribl LogStream last year, we knew pricing was going to be an important component of our value proposition. There’s no question we make management of complex observability systems way safer and easier, but better administration is often hard to quantify in terms of benefits to the bottom line. Nearly all of our prospects are looking for a positive return on their investment in our product. There are several ways they may achieve savings by use of Cribl LogStream: reducing infrastructure and license cost and growth in their existing tooling, minimizing downtime of their tooling due to unexpected traffic or operator error, or avoiding the cost of rolling out yet another agent to thousands of endpoints.
Cribl really wants our model to be as fair and aligned as possible between our customers’ interests and ours. We’ve always felt consumption models are pretty fair. They grow with usage. Infrastructure consumption models by compute seconds are fine, but they put a high burden on being able to auto scale in order to minimize costs. Many data streaming cloud services, like Kinesis, Confluent Cloud, bill on gigabytes throughput. But, cloud native models remove peak daily from the consumption aspect, and so they are actually what you consume. Peak daily consumption is a consumption model, but it’s is totally unfair, because you are paying everyday for capacity you are not consuming. If you buy a 1TB/Day license, but you ingest 750GB/Day on average, 25% of your cost is for unused capacity!
In order to be aligned with our customers, we wanted one of the top use cases, freeing up capacity due to wasted ingest, to be incentivized. We brainstormed, and what we came up with is to charge independently for data input and data output so eliminating waste is rewarded. We call this True Consumption pricing.
With True Consumption pricing, you pay us $.10/GB in and $.10/GB out. If data comes in but does not go out, you pay only for the in but not the out, or half the price. We will have a metered offering available soon in the AWS Marketplace, but right now, this pricing is available in annual term contracts with additional discounting available at higher volumes. For many customers, simply by moving to True Consumption, the model is 20-30% cheaper.
For annual contracts, we can estimate your consumption based on use cases, and true up in year two based on the prior year’s consumption. We will generally negotiate burst capacity into our agreements so customers are free to try out new use cases or exceed their purchased capacity somewhat without having to renew in-year. We will continue to offer Peak Daily Ingestion as well for annual contracts. A single node free version is still available, and will be in perpetuity, at up to 100GB/Day. If you’re interested, please send us an email below.
The rest of this post will walk through some examples of how you might estimate Cribl LogStream’s price in your environment.
Let’s look at a single day. For this customer, we are the core of their observability pipeline. They are bringing in logs and metrics, reshaping, aggregating, and routing to the right destinations. Some sources are trimmed greatly, others are copied and sent to multiple systems greatly increasing the value of the source data. With our distributed management technology, they’ve eliminated their past failures and data loss due to misconfiguration. They are able to dial up instrumentation responsively when issues start arising. They’ve never been able to troubleshoot or dive into security data with this fidelity before.
In this installation, on this day our customer sent in 184 GB to LogStream and LogStream sent back out 296 GB to other destinations. In this case, the customer will pay us (184 * $.10) + (296* $.10), or $48. If this is a pretty typical day, we could estimate this customer would spend about $48*365, or $17,520 a year on LogStream.
In this scenario, we’ve engaged in a Proof of Concept, and we’ve identified a use case where we can easily trim 30% off a high volume, low value log. During the PoC, we saw 1 Terabyte a day on ingest and 700 Gigabytes a day egress on average. The problem we want to solve with True Consumption is to avoid having to overpay for unused capacity. In this case, we quote the customer $62,050 for 620,500 (1000 * 365) + (700 * $.10) gigabytes of annual consumption at $.10 a gigabyte. However, the customer doesn’t know if that’s an exact amount, after all we estimated, and they have a few ideas of new use cases they’d like to deploy this year but they have no idea at purchase what that capacity will look like.
For this customer, to solve our estimation and expansion issues, we offer additional burst capacity in the license for free. For this use case, we offered an additional 50%, or 310,025 gigabytes, of annual capacity to try out new use cases and in case our original estimations were too low. Because there is no in-product enforcement on volumes for annual contracts, even if they customer exceeds the burst capacity, the product will not shut down. All we are asking for is for the customer to engage for an early expansion or to ensure we’re aligned for the contract renewal.
With True Consumption, years 2 and beyond become crystal clear and closely aligned. We look to the prior year, include some new burst capacity for new use cases, and renew at last year’s consumption.
We’re super excited to roll out True Consumption, as we think it represents a model that aligns us closely with our customers. Cribl is a customer’s first company; our success is our customers’ success. We think our new pricing is setting a standard for value in the log ecosystem.
However, for you or your circumstances maybe we aren’t hitting the mark. Our commitment is to ensure every customer feels they are getting a great value, and that Cribl was crucial to the success of their tooling strategy. We can only do that if our price is right. With that, if you think for some reason the price isn’t right, we want to talk. Price should not be an obstacle to becoming a Cribl customer if LogStream will solve your problem.
We look forward to hearing from you! Please reach out to us at firstname.lastname@example.org or join our Community Slack.