Fixing Observability Pricing for the AI Era

Today, we’re announcing new pricing for Bronto. 

TL;DR: 

• $0.10 per GB ingested, $1 per TB searched  – any signal - logs, traces, metrics. No per-host or per metric charges. Retained for 12 months - always hot with sub second search

• It’s designed to be disruptive, simple, predictable, aligns cost with customer value and is >100X more efficient than legacy pricing. 
  

We are addressing the foundational issue our industry has failed to tackle up until now: current pricing models are not fit for purpose - they are completely out of whack from an overall cost perspective, do not align cost with value and are designed to favour the vendor not the customer. They are also hard to understand and wildly unpredictable with teams regularly getting hit with nasty overage surprises.

Cost is and has been the biggest issue for observability customers for the past 15+ years and vendors continue to willfully ignore it. 

Not that anyone needs a reminder…observability costs routinely run at 20–30% of total infra spend. This results in a myriad of issues where teams drop high-volume logs and cut retention to 3, 7 or 15 days. You sample. You archive and rehydrate. You build pipelines to throw data away before it lands. You roll your own observability on open source and inherit all the management overhead. 

The one that stuck with me recently: a company resizing its hosts onto bigger AWS instances just to fight their vendor’s per-host pricing – even though that wasn’t the right architecture for their system. They changed their infrastructure to suit their observability bill. That’s how distorted observability pricing has become.

Instead of tackling this issue head on, vendors continue to add ‘features’ on top of inefficient datastores that are not fit for purpose. The latest wave of capabilities are coming in the form of AI agents and automated workflows. 

Don't get us wrong these capabilities are super powerful and will be a huge help for teams managing and operating complex systems, will reduce MTTR and give better RCA - however if you build AI capabilities on top of fundamentally broken foundations - the cost problem will simply get worse especially as AI workloads will also drive higher and higher data volumes.

Legacy Pricing is Broken

There are really two problems with legacy pricing. The first is that it’s about 10× too expensive no matter how you slice it – vendors are charging dollars per GB ingested and stored. However per GB pricing needs to be at the level of cents per GB to stop teams architecting around the limitations imposed by a high per GB cost i.e. observability needs to drop from ~30% of infra spend to under 5% – low enough that you stop thinking about it and having to engineer around it. 

On top of that, the dollars per GB pricing is typically for days of hot retention - i.e. you are paying through the nose for access to a sliver of your data. Again, retention is at least an order of magnitude off for an AI world where you should be able to do historical analysis using months and years of ‘hot’ data.

The second problem relates to value (or the lack thereof). You pay to ingest and store, so the vendor gets paid whether or not you ever get a cent of value out of the data. You may never log in, you may never search, you may never add alerts, you may never use that data - but the vendor gets paid. We hear it constantly, in logging especially – people describe their provider as a very expensive datastore they never actually use or get any real value from. The model was built for the vendor, not for you.

Enter Bronto - Observability Pricing Built for the Customer, Not the Vendor

Built on top of BrontoDB, our custom built observability datastore, Bronto drives massive efficiencies when it comes to ingesting, storing and analysing your observability data. 

Bronto pricing - cents per GB, not dollars per GB.

We’ve kept it simple enough that your granny could follow it:

• $0.10 per GB ingested – any signal - logs, traces, metrics. No per-host or per metric charges. Retained for 12 months - always hot with sub second search.

• $1 per TB searched – that’s 5× cheaper than scanning the same data on AWS Athena, which runs $5 per TB.

With Bronto if you ingest your data and never search it you pay very little. You only pay more when you get more value by searching across more data. Our incentives line up with yours, not against them. It’s that simple. No more expensive underused datastores.

“So how much search will I need?” For entry level plans, you won’t have to think about it as they come with search built in and are uber generous (for more details see below).

For enterprise plans, usual devops use cases at scale typically come in under a combined ingest and search cost of $0.20 per GB - but you can also sign up for a free trial to verify exactly where you land. 

Bronto is 100x - 1000x More Efficient

We’re in the corner of startups, solo builders, and teams building something new - so our entry plan is uber-generous (for all your logs, traces and metrics) and starts at $25/month for 1TB ingested with 20× search, at 12-month retention – roughly $0.025 per GB, for any signal, with no per-host costs.

Datadog runs about $2.60 per GB for 30-day retention. Putting real numbers on it: a team ingesting 1TB of data a month might pay around $2,600 with Datadog for 30 days retention, versus $25 with us. That’s 100× cheaper and with >10× the retention on top, it works out to be roughly 1000× more efficient. Note that doesn’t even get into cost explosion issues around things like high cardinality metrics.

At larger volumes Bronto charges $0.10 per GB ingested and $1 per TB searched  – for a typical DevOps profile – that comes in around $0.20 per GB (all in) with 12 months retention, against Datadog’s $2.60 per GB at 30 days (assuming ~1KB events + 100% log indexing). Roughly 100× more efficient.

Simple. Predictable. No surprises.

• Simple – one per-GB cost for any signal ingested, one per-TB cost for what you search. >90% more cost-effective than the traditional vendors, and you get 12 month retention by default.

• Predictable – step into an entry plan with uber-generous limits and you’ve likely got a ton of headroom, so you’re not watching costs creep. For larger plans use the free trial & usage dashboard or talk to our team.

• No surprises – built-in usage tracking and alerting means no end-of-month shocks. And we never block you when you hit a limit – we’ll warn you if something looks amiss so you don’t get caught out, but the data keeps flowing.

Bronto Delivers

All your data in one place. Full coverage with no blindspots – the platform efficiency and the pricing mean you can keep everything. Always-hot data at 12-month default retention. Seamless cross-correlation on any signal – go from a log to a trace to a metric without hopping between three separate datastores (e.g. Prometheus to Tempo to Loki, each limited and disjointed in its own way). And your AI on top, via our MCP, Bronto Vibe or Bronto’s built-in investigation capabilities.

Try Bronto Today

Spin up a free trial and run the numbers against your own data. Read the full details at bronto.io/pricing.