Horizon Media · From Noise to Signal
Horizon Media · Challenge 1

See what you’re paying for — and what you can stop storing.

Customer value: A clear baseline of noisy vs useful metrics before the next renewal conversation.

Horizon Media ingests OTel from cloud, on-prem, and direct sources. Most teams discover 10–30% of metrics never drive an alert or dashboard — but still hit the bill.

YOU GET

Honest ingest picture

Scrape churn, debug routes, and high-cardinality labels — named and measurable.

YOU GET

Signal vs noise ratio

What leadership actually monitors: errors, requests, heartbeats — vs everything else.

YOU GET

One stream to discuss

Same PromQL/OTel pipeline — no “rip and replace” story for your SRE team.

metrics-generic.otel-default
Why this matters to the CIO
Before you optimize
Three observability tools · one metrics bill Paying for data nobody queries Audit first → then drop with confidence Outcome: defensible % to cut at renewal
Lower storage cost Same alerts
In the lab: Audit metrics-generic.otel-default in Streams — name your top noise sources.
elastic
Horizon Media · Challenge 2

Cut the bill at ingest — keep what matters for Sev-1.

Customer value: Immediate ingest reduction without losing login, playback, or billing signals.

Streams lets you drop debug traffic, downsample long-range charts, and keep compliance history via federated sources — policy at the stream, not in every dashboard.

YOU GET

~10% off the bill now

Drop /debug/pprof and scrape churn before it’s stored.

YOU GET

Federated long-term store

Compliance and historical trending via federated data sources — without copying everything into observability.

YOU GET

Months of trends, not months of spend

Elastic Serverless is ~25% the cost of Datadog — long-range signal without breaking the budget.

Three levers · one conversation
Drop · Federate · Downsample
All OTel Stream rules Signal only
Same PromQL ~25% of Datadog
In the lab: Apply Drop + lifecycle on your stream — capture ingest ↓ % for your cost story.
elastic
Horizon Media · Challenge 3

A TCO story leadership can repeat in the hallway.

Customer value: PB/TB math + measured savings — not a generic “observability is expensive” slide.

Connect Drop and federated long-term stores to dollars: at 1 PB/month, dropping a quarter at the stream is 250 TB/month off the bill. You leave with numbers, not adjectives.

YOU GET

Measured savings

Real noise vs signal counts from your stream — not vendor estimates.

−10–25%
YOU GET

Why it matters to you

“We pay for signal, not scrape churn — Elastic Agent into Serverless, ~25% Datadog TCO.”

TCO
YOU GET

Path to production

Consolidation proof today → adoption velocity and agentic ROI on the same stack.

Leadership summary
From the savings dashboard
"Elastic Agent ships OTel to Serverless. We dropped [X]% noisy telemetry at ingest — errors stay full fidelity for Sev-1." Horizon — Streams savings dashboard
Renewal-ready
In the lab: Open the savings dashboard — deliver one Drop %, one retention idea, one PB/TB example.
elastic