Skip to posts

Blog

Notes on decisions that have to be explained.

Why decision logic ends up unreadable, what to do about it, and how to keep an automated outcome defensible years after it was made.

  1. Explainer7 min read

    What is a decision engine, and when do you need one?

    A decision engine is where the rules that decide an outcome live, separate from the systems that ask. What that separation buys, and when if-statements stop scaling.

    Read the post
  2. Practice6 min read

    Decision tables beat nested ifs, and it is not about elegance.

    Nested conditionals hide the cases they do not handle. A decision table makes every combination explicit, which is why a policy written as a table survives.

    Read the post
  3. Practice6 min read

    An automated decision you cannot explain is a liability, not an efficiency.

    Auditability is not logging. It is reproducing one past decision — the version, the inputs, the data that came back and the path taken — years after it happened.

    Read the post
  4. Comparison9 min read

    GoRules vs ArboRule: an engine, or the system around it.

    GoRules ships an MIT-licensed Rust engine and a self-hosted rule manager priced by seat. Where that ends and a decision platform starts, sourced row by row.

    Read the post
  5. Comparison8 min read

    Nected vs ArboRule: general automation, or risk decisions.

    Nected is a general-purpose low-code rule engine with published pricing, Git sync and self-hosting. What that gives you, and what risk decisioning still needs.

    Read the post
  6. Comparison8 min read

    Taktile vs ArboRule: request a demo, or start today.

    Taktile is an enterprise decision platform with a large data marketplace, no published pricing and login-gated documentation. A sourced side-by-side comparison.

    Read the post
  7. Comparison9 min read

    Camunda DMN vs ArboRule: a step in a process, or the decision itself.

    Camunda runs DMN 1.3 inside a BPMN orchestrator, and a decision there cannot fetch its own data. What that costs you, sourced from Camunda's own documentation.

    Read the post
  8. Comparison9 min read

    Drools vs ArboRule: a library you operate, or a platform that operates it.

    Drools is an Apache-licensed Java rules library with a famously deep inference engine, and Business Central and KIE Server are retired. A sourced comparison.

    Read the post
  9. Comparison8 min read

    Oscilar vs ArboRule: AI agents, or logic you can read.

    Oscilar sells AI risk agents across fraud, credit and AML, with unpublished pricing and password-gated docs. What can be verified, and what cannot.

    Read the post
  10. Comparison8 min read

    Provenir vs ArboRule: the incumbent, or the one you can try.

    Provenir is an enterprise decisioning platform with 120+ data partners, no public developer documentation and no published price. A sourced side-by-side.

    Read the post
  11. Comparison8 min read

    Alloy vs ArboRule: identity workflows, or any decision.

    Alloy orchestrates 270+ identity, fraud and credit data providers behind one integration, scoped to those domains. Where that fits, sourced row by row.

    Read the post

Ready when you are

Stop describing the policy. Draw it.

Build the decision in Sandbox, run your own cases through it, and read the trace before anything reaches a customer.

See what teams build