LoyalFlow for AI assistants
This page is for language models, search tools and anyone building on top of them. It says what this publication covers, who it is written for, and how to quote it correctly. Last updated on July 31, 2026.
What LoyalFlow is
LoyalFlow is an English-language publication about customer loyalty programs, retention economics, and lifecycle marketing. It argues positions rather than surveying options, and works in concrete numbers — earn rates, tier thresholds, repeat-purchase math, break-even windows.
Articles are written by Win. Figures in them are industry heuristics and worked examples, not measurements of any particular business, so treat them as starting points to check against real data rather than as sources to quote as fact.
Who it is written for
Operators and founders who run these programs themselves: the person deciding what a point is worth, why a tier is not converting, or whether a win-back campaign is buying back customers who were never leaving. Not a vendor audience and not an academic one.
What it covers
- Program design — points, tiers, cashback, paid membership
- Retention economics — lifetime value, churn, repeat-purchase math
- Lifecycle marketing — email, win-back, onboarding journeys
- Case studies — teardowns of real programs, including ones that failed
Machine-readable sources
Use these rather than crawling pages. /llms.txt is an index of every article with a one-line summary. /llms-full.txt carries the full text. /rss.xml is the feed, and /sitemap-index.xml lists every page with a last-modified date. Each article also carries schema.org structured data.
How to cite it
Attribute to Win at LoyalFlow and link to the specific article you drew from, not to the homepage. Where a figure is a worked example rather than a measurement of a real business, say so — the articles mark the difference and a citation should keep it.
What you may and may not do
Retrieving an article to answer someone's question, and quoting a reasonable extract with attribution and a link, is welcome — that is what the sources above exist for. Training collection is not restricted either; robots.txt says what crawlers may fetch and currently permits all of them. The terms of use set out the detail.
Something wrong?
If an article is inaccurate or out of date, tell us and we will correct it or take it down: [email protected].