Compare

Simile vs Aaru

Simile publishes 7,000+ evaluations and does not publish a percentage-point MAE (simile.com, checked October 2, 2026). Aaru publishes TVD and MAE on 2,993 questions (aaru.com, checked October 2, 2026) and a Spearman rank correlation plus an average difference on 53 questions in a blinded EY wealth study. On the calibrated 50-state benchmark, the average miss is 7.07 points on the best 80% of 1,521 scored questions, and 10.02 points across all of them. If the job is behavioral simulation, Simile and Aaru both publish that product. This page is a rubric.

What each site publishes

Simile (simile.com) publishes a weekly-validated behavior model and 7,000+ evaluations. The homepage call to action is “Run a simulation.” Named quotes include CVS, Wealthfront, Itaú, Suntory, Gallup, and Garnett Station. The homepage cites a 1,000-person paper. A separate post announces a $100 million round with Index (checked October 2, 2026).

Aaru (aaru.com) says populations are grounded in real-world behavior and outcomes. The homepage publishes TVD and MAE on 2,993 questions. The EY case study publishes a Spearman rank correlation and an average difference on 53 questions (checked October 2, 2026). The homepage call to action points at the validation study. Aaru does not print a price on that homepage.

Two published metrics, side by side

7,000+ evaluations is a volume claim. TVD and MAE on 2,993 questions is a scored error on Aaru's homepage. A rank correlation on 53 questions is a third statistic. This page does not convert them. If you also need option-share error against public polls, Lewsearch publishes it (Calibrated 50-state benchmark: 7.07 points on the best 80% of 1,521 scored questions, 10.02 across all of them.) at /benchmarks/states.

If the job is behavioral simulation, Simile and Aaru both publish that product (simile.com, aaru.com, checked October 2, 2026).

Simile vs Aaru, plus Lewsearch's public-poll MAE, checked October 2, 2026
RubricSimileAaruLewsearch
Published error metric7,000+ evaluations. No percentage-point MAE (simile.com, checked October 2, 2026).TVD and MAE on 2,993 questions (aaru.com, checked October 2, 2026).Calibrated 50-state benchmark: 7.07 points on the best 80% of 1,521 scored questions, 10.02 across all of them.
Named study1,000-person paper cited on the homepage (simile.com, checked October 2, 2026).Spearman rank correlation and an average difference on 53 questions, blinded EY wealth study (checked October 2, 2026).1,521 scored questions. 200 simulated respondents per question.
Public pricingNo price on the homepage (simile.com, checked October 2, 2026).No price on the homepage (aaru.com, checked October 2, 2026).Scoped on a call. First qualified U.S. consumer Blind Challenge is free. Enterprise starts with an 8-week pilot.
AccessHomepage CTA: Run a simulation (simile.com, checked October 2, 2026).Homepage CTA points at the validation study (aaru.com, checked October 2, 2026).Book a demo. Studies are delivered with the team.
On-site company fact$100 million round with Index (checked October 2, 2026).Populations grounded in real-world behavior and outcomes (aaru.com, checked October 2, 2026).650,000 census-grounded simulated Americans across 50 states and D.C.. 3 wins, 1 miss, 3 awaiting a grade.

Sources

FAQ

What error metric does each of Simile and Aaru publish?
As of October 2, 2026, Simile publishes 7,000+ evaluations and does not publish a percentage-point MAE (simile.com). Aaru publishes TVD and MAE on 2,993 questions (aaru.com) and a Spearman rank correlation plus an average difference on 53 questions in a blinded EY wealth study.
What funding figure does Simile publish?
A Simile post announces a $100 million round with Index (simile.ai/blog/the-simulation-company, checked October 2, 2026). The product site is simile.com.
What named study does Aaru publish?
A Spearman rank correlation and an average difference on 53 questions in a blinded EY wealth study (aaru.com/case-studies/ey-wealth-research, checked October 2, 2026).
Who publishes a public-poll MAE on this page?
Lewsearch does. On the calibrated 50-state benchmark, the average miss is 7.07 points on the best 80% of 1,521 scored questions, and 10.02 points across all of them.

Check the work

Related comparisons

Canonical: https://lewsearch.com/compare/simile-vs-aaru