# Asabisoft

> Asabisoft is a data science and analytics consultancy: LTV, churn and demand models, data pipelines and dashboards, and the Shopify stores and SaaS products that use them.

Asabisoft is run by Arif Sait Birincioglu, a data scientist and Python engineer based in Izmir, Turkey, with 13+ years across data science, analytics, engineering and product (Forte, N3TWORK Studios, Jawabkom; founder of URLAudit). Clients are served remotely in the US, Europe and Australia.

## Services

- Data science: LTV, churn and demand forecasting; experiment design and A/B test analysis; economy, pricing and scenario simulation; launch / green-light models; ML and LLM automation.
- Analytics engineering: event instrumentation, pipelines into BigQuery, Snowflake or PostgreSQL, data validators, KPI and unit-economics dashboards (Power BI, Tableau), scraping and GIS data.
- Commerce & product engineering: Shopify builds, apps and migrations; SaaS and internal tools in Python (Flask, Celery, Playwright); Stripe billing.

## Pages

- [Home](https://asabisoft.com/): services, data science track record, selected work, process, about, FAQ
- [Writing](https://asabisoft.com/blog/): articles index

## Articles

- [LTV forecasting with under a year of data](https://asabisoft.com/blog/ltv-forecasting-short-history/): With under a year of history, forecast LTV with a probabilistic repeat-purchase model (BG/NBD or Pareto/NBD for purchase counts, Gamma-Gamma for order value) fitted on calibration data and checked against a holdout period. Report it per acquisition cohort, cap the horizon at roughly twice the history you have, and compare it with a simple cohort revenue curve before anyone uses it to set acquisition budgets.
- [What we check before touching a Shopify theme](https://asabisoft.com/blog/shopify-audit-before-redesign/): Before changing a Shopify theme, audit four things in order: speed (Core Web Vitals and the weight of apps and scripts), tracking (whether purchase and funnel events are complete and match Shopify orders), the checkout funnel (where sessions drop, by device and source) and unit economics (customer value against acquisition cost). Fixing these first tells you whether a redesign is needed at all — and which templates it should touch.

## Contact

- LinkedIn: https://www.linkedin.com/in/asbirincioglu
- Founder: https://arifsaitbirincioglu.com