Kleyr review
An analyst-focused MMM workbench with experiment links and reproducible runs.
Reviewed by iTechGuides Editors · Editorial team · Updated Oct 2026
Kleyr is a web-first marketing mix modeling workbench for analysts and marketing operators who want to build and inspect models directly. It supports CSV uploads with KPI, media, and control tagging, then offers adstock and saturation transforms, OLS or Ridge regression, diagnostics, and channel decomposition. The Starter plan includes one MMM model fit and one geo-lift or A/B experiment, making it a way to evaluate the modeling workflow before moving to a paid tier. It also includes dataset upload, a reproducibility lockfile, an audit log, and EU data residency.
The modeling workflow is the product's main strength. Users can choose geometric or Weibull adstock and Hill, Michaelis-Menten, or S-curve saturation transforms, then apply stepwise variable selection. Diagnostics include Durbin-Watson, Breusch-Pagan, Ljung-Box, VIF, and residual plots. Per-period channel decomposition and response curves help examine model outputs, while run artifacts preserve a dataset snapshot, configuration, lockfile, and random seed. Experiment-derived geo-lift estimates can be attached as priors to later MMM runs, which suits teams connecting incrementality tests with ongoing measurement.
Plan choice shapes how much of the work is manual or AI-assisted. Operator adds manual modeling software, the full Modeler, experiments, Customer Knowledge features, and a monthly model-fit allowance. Analyst adds a conversational AI agent for end-to-end modeling, a monthly agent-task allowance, higher fit and compute allowances, and priority support. Enterprise adds multiple workspaces, a dedicated success engineer for the first 90 days, a public API and webhooks, negotiated allowances, SSO, and custom SLAs. Shopify, Klaviyo, Meta, Google, TikTok, and GA4 are listed, but described as post-v1 rather than evidenced as available connectors. Choose Kleyr for analyst-led OLS/Ridge work and reproducible experiments; teams requiring available integrations or Bayesian modeling now should consider alternatives.
Kleyr pros and cons
- Where it wins
- Supports adstock and saturation transforms with OLS or Ridge regression.
- Includes diagnostics, channel decomposition, and response curves.
- Reproducible artifacts capture data snapshots, settings, lockfiles, and seeds.
- Where it doesn't
- Current model workflow is traditional OLS/Ridge MMM; Bayesian modeling is planned.
- Listed service connectors are described as post-v1, not evidenced as available.
- Manual modeling and AI-assisted modeling are split across paid tiers.
Kleyr fact sheet, pricing and score →
Advertiser disclosure: iTechGuides is reader-supported. Vendors can pay for top positions in our rankings and for a place on other products' pages, and we may earn a commission when you click some links. How we rank.
Last updated · How we research and update