Skip to content

README.md

Latest commit

 

History

History
84 lines (55 loc) · 5 KB

README.md

File metadata and controls

84 lines (55 loc) · 5 KB
Raw

Research: Replacing TaxJar with an Internal Solution

Date: April 2026 Goal: Evaluate whether we can stop paying for TaxJar (taxjar.com) and build an internal sales tax compliance tool for our Amazon seller business.

TL;DR

We pay $1,128/year for TaxJar (Basic 5K at $528/yr + $600/yr flex fees). This jumps to $1,599/year after October 2026. We are NOT using AutoFile. Boss uses TaxJar only to "calculate the info needed to file state taxes." His verdict: "the cost does not justify the info we're getting."

The situation is simpler than we thought:

  • We file in only 5 states: WA (quarterly), NC (quarterly), GA (annual), OH (semi-annual), MN (annual)
  • We file $0 returns in ALL states -- Amazon and other marketplace facilitators (eBay, Walmart) handle everything
  • Shopify doesn't collect tax, but volume is too small to matter
  • We already have Amazon SP-API access (sunk cost)

Bottom line: We're paying $1,128/yr (soon $1,599/yr) for a report viewer that tells the boss what numbers to put on $0 returns across 5 states. A simple Python script pulling Amazon SP-API data replaces this entirely at $0 additional cost.

Research Index

Understanding the Current State

# Document Description
01 TaxJar Overview What TaxJar does, core features, pricing tiers, and known pain points post-Stripe acquisition
02 TaxJar API Reference Complete API endpoint catalog -- the surface area we'd need to replicate
03 Amazon Integration How TaxJar connects to Amazon, data flows, and what Amazon handles vs. doesn't

Understanding the Problem Space

# Document Description
04 US Sales Tax Complexity 12,400+ jurisdictions, nexus rules, product taxability, sourcing rules -- why this is hard
05 Tax Rate Data Sources Free/paid APIs, SST government files, open-source options (spoiler: none are production-ready)
06 Amazon Marketplace Facilitator Key finding: Amazon handles 90%+ of the problem. Remaining obligations are filing and nexus tracking

Building the Replacement

# Document Description
07 Replacement Approaches Four approaches evaluated: Rate API, Amazon Data, State APIs, Local DB -- with comparison matrix
08 Recommended Architecture 3-tier hybrid design: Amazon SP-API + TaxCloud + local reference DB, with component breakdown
09 Python Ecosystem Libraries for SP-API, tax rates, PDF generation, dashboards, scheduling -- the full tech stack
10 Cost Comparison Side-by-side cost analysis: TaxJar vs. all DIY options, break-even analysis, risk factors
11 Continuation Prompt Ready-to-paste prompt for the next session after getting boss's answers

Key Findings

1. Amazon Already Does Most of the Work

Since 2023, Amazon collects and remits sales tax in all 45 states as a marketplace facilitator. For Amazon-only sellers, tax calculation is a solved problem.

2. The Real Gap is Filing and Nexus Tracking

  • Most states still require sellers to file returns (even $0 returns) in states where they have nexus
  • FBA inventory in 25+ states creates physical nexus with filing obligations
  • Amazon does NOT file your returns or track your nexus

3. TaxCloud Fills the Filing Gap for Free (in 24 States)

TaxCloud is an SST Certified Service Provider offering free calculation, filing, and remittance in 24 SST member states. Non-SST states cost $39-$45 per filing.

4. Our Actual Cost is Lower Than Expected -- But Rising

We pay $1,128/year currently (not $7,500-$13,000 like heavy AutoFile users). But the October 2026 price hike pushes us to $1,599/year for the same service. The cost bar to beat is modest, but a DIY solution pulling Amazon reports + basic nexus tracking could eliminate this cost entirely.

5. No Open-Source Tax Engine Exists

There is no production-ready open-source alternative. The data maintenance burden (681+ rate changes/year) makes this impractical for community projects.

Recommended Next Steps

  1. Audit current TaxJar usage Done: $1,128/yr, Basic 5K plan, no AutoFile
  2. Determine channel mix Done: Amazon (primary), eBay, Walmart, TikTok, Shopify. All marketplace facilitators except Shopify (negligible volume).
  3. Identify filing states Done: 5 states — WA (quarterly), NC (quarterly), GA (annual), OH (semi-annual), MN (annual). All $0 returns.
  4. Clarify pain point Done: "Cost does not justify the info we're getting"
  5. Build a simple Python tool that pulls Amazon SP-API tax data and generates the numbers the boss needs for $0 filings in 5 states
  6. Cancel TaxJar once the tool is verified against a quarter's worth of data