Datadog
Each of these is a real decision Datadog is working through, written so you can make the call yourself — incomplete information, someone who pushes back, no answer key.
Datadog provides a cloud-based SaaS platform that collects, analyzes, and correlates telemetry and operational data for observability and security across cloud, hybrid, and distributed environments. Over thirty products share one platform, including always-on alerting, metrics visualization, logs, application tracing, and security solutions Cloud SIEM, Cloud Workload Security, and CSPM. It has 600+ integrations and serves development, IT operations, platform engineering, security, DevOps, data, and business teams, handling trillions of data points per day for tens of thousands of companies. It works with distributors Carahsoft and Vertosoft on GSA Schedule and SEWP contract vehicles. In Japan, a local data centre and over 200 local staff localize its products and services for Japanese customers. Its SDKs deliver APM and are the platform for delivering all its service and app based products. Its Event Platform is built around Husky, a distributed, petabyte-scale columnar event store powering dozens of its most popular products. Its Kubernetes platform spans 100,000+ virtual machines across multiple cloud providers. The Datadog Agent is its flagship open-source project.
Your first one is free. No card.
41 live right now.
The rest
Cluster APM feedback before roadmap reviews lose the evidence
Intelligent Troubleshooting feedback is being read as separate anecdotes, even when customers describe the same investigation dead end. You must weigh faster synthesis against keeping every theme reviewable back to the customer evidence.
Choose the next campaign test without losing the buyer signal
Commercial briefs are being reopened because campaign results and buyer discussions point to different next moves. You have to decide what AI can recommend, what evidence it must show, and where strategist judgment stays final.
Decide what AI ranks and what engineers confirm when traces disagree
Recurring failures send engineers through traces, profiles, and changes without agreement on where the chain began. You must decide what AI can assemble and rank, and what evidence still requires an engineer's judgment.
Trace Kafka lag before competing theories delay the release
Consumer-lag investigations produce conflicting explanations across Kafka, Kubernetes, traces, and Redis. You must decide what AI assembles, what engineers verify, and how the team runs it afterward.
Choose a guided scenario for buyers instead of broad AI content
Existing broad AI content engages fewer than 2% of Commercial audiences. You must weigh broad discovery against a narrower workflow story that buyers can actually place in their day.
Decide on the sponsorship before Friday's nonrefundable deadline
The attendee export has a 42% duplicate rate and conflicting account matches. You must weigh exclusive market access against pipeline evidence that can withstand strict review.
Lift workshop attendance while one event budget remains
Attendance fell from 58% to 31% after the confirmation path changed. You must weigh recovering engaged registrants against buying more top-of-funnel reach.
Decide how to show where trace context is lost without overstating gaps
Partial traces leave engineers manually comparing proxy headers and service spans to locate lost context. You have to weigh a simple guided view against the risk of overstating incomplete evidence.
Commit sampling defaults while rare error traces remain at risk
Replay data shows the proposed default cuts ingestion but drops rare confirmed error paths. You have to weigh capacity relief against evidence engineers cannot recover after an incident.
Fix the slow path in Go middleware without breaking request context
Affected Go services add 8 ms of p99 overhead after the middleware release. You have to weigh a fast compatible patch against the risk of breaking request context.
Trace journey drop-offs into service owners without runaway query costs
Evaluation customers can spot a journey drop-off but cannot connect it to a responsible technical surface. You have to weigh a useful cross-product investigation against access, cost, and an understandable first workflow.
Set rollback guidance before responders lose trust in it
Two evaluation customers received rollback guidance tied to changes that were not responsible for the incident. You must weigh broad incident coverage against guidance responders can safely act on.
Guide first DDSQL queries after a passing connection test
Only 41% of connected workspaces reach a first DDSQL query within one day. You must weigh clearer guidance against preserving the access controls that make warehouse connections safe.
Win shared owners for three evaluations without blanket discounts
Security-led prospects are stalling because no shared evaluation owner exists. You must weigh a focused security entry against the broader engineering proof needed to expand it.
Close a $1.8 million expansion while finance demands a fixed ceiling
A $1.8 million expansion is stalled because finance wants a firm ceiling while security rejects less Cloud SIEM coverage. You must weigh predictable spend against the coverage that platform and security leaders will defend.
Win the shortlist before AWS becomes the easier consolidation choice
A $450,000 opportunity is stalled because the prospect cannot agree which tools the platform can replace. You must weigh a broader consolidation case against security’s need to preserve its workflow.
Choose the outcome links for agent traces without building analytics
Design partners can inspect agent behavior but cannot connect it to whether customers completed the intended task. You must weigh a useful shared outcome signal against a first release that stays narrow and verifiable.
Commit the release slot to injection review before engineering begins
A design partner cannot reconstruct a suspected prompt-injection action from its trace and evaluations. You must weigh a defensible investigation path against broader evaluation coverage in the only committed release slot.
Clarify agent handoffs so engineers stop reopening the wrong traces
Engineers investigating failed handoffs are filtering into the initiating agent, not the agent that made the failing tool call. You have to weigh a clearer ownership signal against preserving the trace filtering customers already use.
Connect telemetry spend to SEV1 readiness without a price fight
Several enterprise customers are being asked to reduce telemetry spend without agreement on what protects SEV1 investigation. You have to weigh a repeatable FinOps programme against the risk of oversimplifying each customer’s incident workflow.
Unify the renewal case while telemetry costs face a hard deadline
A renewal is frozen because DevOps, security, and procurement disagree on the value of broader telemetry. You must weigh incident visibility against a commitment the customer cannot easily unwind.
Restore request tracing for on-call within a 5% spend ceiling
APM usage fell after a cost review, and on-call engineers lost request paths across two services. You have to weigh a smaller telemetry footprint against the evidence responders need during incidents.
Forecast agent cost before routine traces become expensive
A successful procurement agent workflow is consuming twice its expected token spend on some days. You must weigh early warnings against noisy alerts and added ingestion cost.
Prove rejected requests have owners before span billing changes
Rejected LLM requests lose their calling source during traffic spikes, blocking investigation and source-level cost analysis. You must weigh attribution coverage against false labels and permanent span-volume exposure.
Fix the judge so it fails answers built on empty tool results
Passing scores are labeling incomplete agent answers as healthy after tool calls return no usable result. You must weigh stronger failure detection against false alarms, latency, and evaluation cost.
Rank early alerts so multi-service incidents surface sooner
Forty-six of 80 reviewed incidents showed an early alert more than 15 minutes before cross-service impact was recognized. You have to weigh useful early ranking against the risk of assigning confidence telemetry has not earned.
Decide whether to commit to a one-way migration before release closes
Thirty-one confirmed account-takeover cases were missed last month. You have to weigh a recall gain against uncertain false-positive volume and an irreversible migration.
Cut the noise from nightly maintenance without missing real threats
Maintenance-window alerts rose from 42 to 214 per week. You have to weigh a faster noise reduction against preserving real anomaly recall.
Choose explainable evidence over broad coverage when alerts fire
Responders are opening four or more telemetry views before finding a plausible recent change. You must weigh a broad first workflow against evidence people can inspect and trust.
Choose a REDAPL rebuild or targeted fixes while tenant counts disagree
A consistency monitor found contradictory Kubernetes ownership links, but affected-tenant estimates range from 19 to 63. You must weigh a hard-to-reverse rebuild against customer write continuity and incomplete evidence.
Fix lookups of high-fan-out requests so p95 drops under 800 ms
p95 traversal latency reached 3.4 seconds for a known high-fan-out request pattern. You must weigh faster graph reads against the ingestion headroom shared by every tenant.
Scope a check of telemetry recovery while fault limits stay unclear
Customers lack a safe way to test whether telemetry survives a regional dependency failure. You have to weigh useful recovery evidence against the risk of customer-operated disruption.
Rotate credentials within 48 hours without breaking ingestion
A confirmed exposure may span multiple regions, while rotation could interrupt high-volume telemetry ingestion. You have to weigh a shorter exposure window against a verified path for incident evidence.
Fix the stale policy on 14 services instead of replaying the fleet
Fourteen services are showing a 6.8x increase in authorization failures after one distribution update. You have to restore access quickly while keeping the shared rollout path safe.
Rank the first follow-up for slow requests without overriding engineers
Engineers are switching among traces, profiles, logs, and metrics before finding a useful next step. You have to decide what guidance deserves ranking and how to prove it helps without taking control away.
Hold the automatic trigger until blinded review shows 90% attribution
The pilot finds the right deployment in 84% of reviewed incidents, but false leads can redirect outage response. You have to weigh faster investigation starts against the cost of a wrong attribution becoming operational guidance.
Recalibrate error clusters after unrelated timeouts flood investigations
Error-cluster precision fell from 91% to 76% after a traffic-pattern change. You have to restore useful grouping while protecting recall for real incident patterns.
Decide a token view for traces instead of a cost suite
Design partners cannot connect rising token use to the request paths responsible for it. You have to weigh a focused investigation view against the pull toward a broader cost-management product.
Gate root-cause guidance while incidents still need answers
Three high-severity investigations named the wrong service as the likely cause. You have to weigh incident speed against a trust boundary that cannot be casually reversed.
Fix traces in Java services within 30 days
Complete traces for affected Java services have fallen from 88% to 71%. You have to weigh a fast onboarding change against the risk of confusing existing instrumentation.