How the writing here gets made.
Our articles are drafted with AI to a brief a person wrote, then edited against fixed standards before they publish.
How an article gets written.
Four steps, in the order they happen. None of this is hidden, because a family deciding whether to trust what they are reading deserves to know where it came from.
A person decides the article should exist
Nothing here is written because a keyword tool suggested it. We keep a written map of the questions families actually arrive with, and every article is one of those questions. A person picks the subject, decides who it is for, and lists what the piece has to answer before a single word is drafted.
The draft is written with AI
We use AI models to write the first draft against that brief, and we would rather say so than let you work it out. The model is given the questions to answer, the reading level, the things it may not claim, and the pages it should link to. It is a drafting tool working inside instructions a person wrote, not an author with opinions of its own.
A separate editing pass rewrites it
Every draft then goes through an editing stage that checks it against the standards below: does it answer the question in the first hundred words, is anything asserted that cannot be supported, is the tone right for someone reading this at the worst moment of their year. Weak sections get rewritten or cut. Automated checks run last and hold back anything that fails them as an unpublished draft.
A person is answerable for what went out
We publish on a schedule, so a person does not read every article in the minutes before it goes live. Abdullah Nabeel, the founder and the editor of this blog, is answerable for what went out, and the responsibility for it does not move onto the software. If something here is wrong, it is ours to fix, and the correction path below is the fastest way to make that happen.
What every article has to meet.
What every piece holds to
Facts that change carry a date
Funeral costs, benefit amounts, filing deadlines, and state rules all move. Wherever an article gives one, it says when that figure was accurate and points you at the office or agency that holds the current version. A number with no date attached is a number you cannot use, so we do not publish them that way.
Our tools show their sources
The cost estimators and the funeral assistance finder follow the same rules as our articles. Every price, benefit amount, cap and deadline they use comes from a named source, usually the agency's own page, a statute or a published price study. Each figure carries the date we last checked it, and we review them on a schedule, sooner when a rate is known to change, such as the VA's burial allowances each October.
Written for someone in the middle of it
People read these pages days after a death, often at two in the morning. So the writing stays plain, gets to the answer quickly, and never uses fear or urgency to move anyone along. We do not lecture families about grieving correctly, and we do not treat a loss as an opportunity to sell to someone.
We say where our knowledge ends
Articles about probate, death certificates, benefits, or a body's care explain how the process usually works and then tell you who to ask about your own situation, whether that is a funeral director, an attorney, a doctor, or the agency itself. General information is not legal, medical, or financial advice, and we do not dress it up as any of those.
What none of them ever does
Nothing is invented to fill a gap
No made-up statistics, no made-up studies, no made-up sources. Articles written since September 27, 2026 link to the sources behind their facts. If we cannot stand behind a number, it does not go in the article, and a sentence gets rewritten around the gap instead. A confident-sounding figure with nothing behind it is the single most common failure of AI-written content, and it is the one we watch hardest for.
Quotes belong to people who said them
We do not attribute words to a funeral director, a grief counsellor, a family, or any named person unless those words are genuinely theirs. Where an article shows sample wording for an obituary, a eulogy, or a condolence note, that wording is written as an example and labelled as one. The names in our sample memorials are fictional.
No one pays to appear here
We run no sponsored posts, no paid placements, and no affiliate links. Articles do link to our own tools, such as the obituary writer or the memorial builder, where they genuinely help with the task at hand, and those links are ours and are obvious. If a mention only exists to sell something, it does not survive the edit.
Found something wrong? Tell us and we will fix it.
Email support@lovekept.co with the address of the page and what is wrong with it. You do not need to prove anything or explain who you are. A pointer to the sentence is plenty.
We check it against the source. Anything we cannot support comes out the same day we confirm it, and an article that is wrong enough to mislead someone is unpublished rather than left up while we work on it. When a correction changes what an article actually says, we note the change and the date at the foot of the piece instead of quietly editing it.
The same address works for anything else about this content: a question we should be answering, a claim you think is outdated, or a piece that reads coldly to someone who has just lost someone. That last one matters to us as much as the facts do.
Who is behind the byline.
Articles are published under the Lovekept Editorial Team. That is a team name rather than one writer's name because more than one person and more than one tool has a hand in each piece, and inventing a fictional author with a fictional biography would be its own small lie.
Lovekept is independent, based in the United States, and has been running since 2026. No funeral home, cemetery, or aftercare provider owns a piece of it or has any say in what we publish. A real person replies at support@lovekept.co
For questions about using the product rather than about what we have written, the help center covers setting up and running a memorial page, and our about page explains how the platform itself works.
The editor.
Abdullah Nabeel, the founder of Lovekept, is the editor of this blog. That name is on every article as its editor, and the editor is answerable for what the blog publishes: what the articles say, what they get wrong, and how quickly a mistake gets put right.
Being the editor does not mean reading each article before it goes live. Articles publish on a schedule, as the steps above explain. It means the responsibility for them sits with a named person rather than with the software that helped draft them.
Written to be useful on the worst week of your year.
If an article here helped, the memorial itself is free to build and takes about ten minutes. If one of them let you down, the correction address above reaches a person.
- No invented facts or sources
- Changing figures carry a date
- Corrections made promptly