# Async AI workflows with cached retries

> Push heavier AI jobs into asynchronous workflows and cache the results so a retry does not force the model to redo expensive work.

- Canonical HTML: https://growth.iangoh.com/growth-ideas/async-ai-workflows-with-cached-retries/
- Source: [newsletter.posthog.com](https://newsletter.posthog.com/p/what-weve-learned-about-building)
- GrowthDex source hub: [PostHog Newsletter](/sources/posthog-newsletter-newsletter-posthog-com-2/)
- Last checked: 2026-05-26
- Rarity: rare
- Budget: medium
- Channels: Product, Retention, Operations
- Stages: ai products, operations, latency, reliability


## Why this can grow

Some AI jobs are too large to fake as instant. Long summaries, extraction jobs, and cross-record reasoning often need more time than a normal request-response loop can hide. An async workflow sets the right expectation and keeps the interface honest. Caching the expensive step means the team can recover from failures without asking the user to wait through the whole job again, which protects trust and keeps operational costs from growing every time the workflow hits a rough edge.

## Ian's take

From scaling consumer platforms across MENA and Southeast Asia, my default is to distrust growth work that only looks good in a slide. My bias is to treat this as a small market test first. Make the audience narrow, make the promise concrete, and let the first real response decide whether it deserves more work. I would run it small enough to learn quickly, then only scale the parts that real users repeat, save, reply to, or buy from. For this tactic, I would watch one clear growth signal before putting more time or budget behind it.

## Action plan

1. Define one narrow startup segment where async ai workflows with cached retries can create a measurable lift.
2. Turn the tactic into one offer, page, campaign, or workflow for the Product and Retention channel.
3. Use the evidence from newsletter.posthog.com to set the first version of the message, format, and audience.
4. Launch a small test for 7 to 14 days with one success metric: one measurable growth signal.
5. Review the result, keep the winning message, remove weak variants, and turn the learning into a repeatable growth playbook.

## Source-backed example

PostHog uses Temporal workflows and Redis caching for heavier Max AI tasks so long-running processing can finish outside the main interaction while retries reuse previous work.

## Adjacent tactics in the same lane

- [Task-based model routing for AI speed](/growth-ideas/task-based-model-routing-for-ai-speed/) - same source, 2 shared channels, 2 shared stages
- [Workflow-first AI demand validation](/growth-ideas/workflow-first-ai-demand-validation/) - same source, 2 shared channels, 1 shared stage
- [Weekly traces hour for agent quality](/growth-ideas/weekly-traces-hour-for-agent-quality/) - same source, 2 shared channels, 1 shared stage
- [Uncertainty, source, and progress cues in AI UI](/growth-ideas/uncertainty-source-and-progress-cues-in-ai-ui/) - same source, 2 shared channels, 1 shared stage

## Read GrowthDex essays

Browse the plain-English essay index at [GrowthDex Blog](/blog/).

## Related GrowthDex essays

- [The AI feature only feels smart after the first useful minute](/blog/the-ai-feature-only-feels-smart-after-the-first-useful-minute/) - ai products, activation, product UX

## Advisory

If you want help turning this into a working growth system, Ian Goh offers advisory at https://iangoh.com/advisory.