All workflows
Operations

News and filing alert triage

High-frequency classification and summarization of news and SEC filings into structured alerts routed by materiality tier.

News alert triage on a Slack + Teams + Zapier stack: 400 runs/day, 61% hidden in prompt overhead, coverage context, and connector routing. Firedog tracks cost-per-tier and connector drift.

Model 15%Prompt 24%Orchestration 11%Retrieval 2%History 39%Connectors 9%
ComponentWhy it is unattributedcalls$ / run
Model · 15%Classification and summarization (Claude Haiku 4)Each news item or filing alert is sent to Haiku 4 for materiality classification (5-tier) and a 150-token structured summary.tokens in: 2,200 · tokens out: 180Visible on the invoice.1$0.003
Prompt · 24%Classification schema and entity metadataA 4,800-token prompt carries the materiality taxonomy, covered-name list, and 6 few-shot examples. Sent on every call with no caching enabled.tokens in: 4,800 · tokens out: 0Visible on the invoice.1$0.005
Orchestration · 11%Ingestion, dedup, and routing pipelineA dedup agent checks the item against a 24-hour seen-items cache; a routing agent sends Tier-1 alerts to a downstream escalation model. Both are billed calls.tokens in: 1,400 · tokens out: 200Dedup and routing agents are considered 'infrastructure' and billed to the platform cost center, not the signal workflow budget.2$0.002
Retrieval · 2%Entity resolution and prior-alert lookupEach item triggers an entity resolution query to map freetext company names to internal tickers, plus a prior-alert lookup to detect duplicates.tokens in: 900 · tokens out: 0Entity resolution queries are served by an in-house NER model whose inference cost is allocated to the infra team, not the trading workflow.2$0.001
History · 39%Coverage context re-injectionFor Tier-1 items, the last 3 alerts on the same ticker are prepended to provide trend context. This multiplies input tokens for the highest-value subset of items.tokens in: 7,500 · tokens out: 0Coverage context is injected by the alert-enrichment service; it shows up as increased prompt length in the model bill but is never attributed to this workflow.1$0.008
Connectors · 9%Slack, Teams, Zapier, and Salesforce routing connectorsEach material item is routed to Slack and, for Tier-1, cross-posted to Teams, logged to Salesforce, and handed to downstream Zaps. Routing carries a small payload-formatting call per destination; at 400 runs/day the connector volume dominates the marginal cost of the cheap classification model.tokens in: 700 · tokens out: 120Slack/Teams posts, Salesforce logging, and each Zapier task bill to their own platform plans; the routing volume scales with alert flow, not headcount, and never reaches the triage workflow budget.2$0.002
At $0.0205/run and 400 runs/day, news triage looks trivially cheap - until you realize 61% is uncached prompt overhead, routing infrastructure, coverage context, and Slack/Teams/Zapier connector traffic that scales with alert volume, not analyst headcount.
$0.021
cost per run
$21k
per year
61%
unattributed today
$8.20
per user, per day
Scenario: Event-driven companies, 300-name universe. 10 active users, 400 runs/day, 250 working days. 1 documents/run. Avg doc: 1,800 tokens.

Default model: claude-haiku-4

Trigger: Continuous stream during market hours: ~200-800 raw items per hour depending on market conditions and covered-name universe.

Who runs it: Trading desks, event-driven companies, and risk teams at large asset managers. Volume is high; latency requirements are strict (sub-30s); models must be fast and cheap. Alerts are routed through Slack and Teams, logged to Salesforce, and handed to existing Zapier flows, so the connected stack carries every material item downstream.

What Firedog shows

  • Total daily and monthly triage cost with per-run breakdown, updated in real time.
  • Cost-per-alert by materiality tier, showing that Tier-1 items (context-enriched) cost 3x Tier-4 items.
  • Projected annual saving from caching the classification schema prefix across 400 daily runs.
  • Connector cost per routed alert across Slack, Teams, Salesforce, and Zapier, showing how routing dominates at high volume.

Decisions this informs

  • Cache the 3,800-token classification schema prefix; at 400 runs/day and $1/M input, caching saves ~$0.49/day - $122/yr at 90% cache hit rate.
  • Limit prior-alert context injection to Tier-1 and Tier-2 items only; Tier-3 and below do not benefit from trend context.
  • Batch Tier-4 and Tier-5 items with a 60-second delay window to reduce per-item call overhead.
  • Route only Tier-1 and Tier-2 items to Slack/Teams/Salesforce; batch lower tiers into a single digest Zap to cut per-item connector traffic.

Connected systems

  • Slack · Collaboration · Salesforce (Slack)
    Material alerts are routed to desk Slack channels by materiality tier.
  • Microsoft Teams · Collaboration · Microsoft
    Tier-1 escalations are cross-posted to the risk team's Teams channel.
  • Zapier · Automation
    Existing Zaps handle AP/AR-style routing of lower-tier items; the triage run orchestrates on top of them.
  • Salesforce · CRM
    Material alerts are logged against covered-name records in Salesforce.

Understand every AI dollar.

Book a demo to see how Firedog attributes every AI request across your organization, and what your leadership team can do with the data.