--
title: "Choosing a Deletion Scope: Everything, By Date, or By Risk"
description: "Scope decides the result and how reversible it is. Four selection strategies, their failure modes, and the three checks to run before starting."
tags: ["privacy", "deletion", "twitter", "cleanup"]
canonical_url: https://digital-footprint-health.shop/blog/deletion-scope-selection
Most deletion advice focuses on which tool to use. The decision that actually determines the outcome is the one made before any tool runs, which is how much you are removing.
Scope decides two things at once: the result, and how reversible the whole operation is. That combination makes it the highest-consequence choice in the process.
1. Four selection strategies and what each one is for
Scope decides both the outcome and how reversible the cleanup is, which makes it a more consequential decision than which tool you use.
Deleting everything is the only option with no partial state. It is the right endpoint if you are leaving the platform and the account will not be used again, because exposure drops to zero in one pass and there is nothing left to think about.
The cost is that normal content goes with it. The timeline empties out, and that is not recoverable from the platform side.
2. When a date range is genuinely enough
Deleting everything fits leaving a platform, deleting by date fits traces tied to a period, and deleting by risk has the best hit rate while depending on a report you trust.
Deleting by date is straightforward to run and easy to split into batches, which is why it stays popular. It handles any goal that is already bound to time: pushing down student-era posts before a job search, or clearing check-ins that carry an old address before a move.
It is a poor main strategy for risk. One person might post location-heavy content during a period that is not the one they wanted to remove, and there is no way to know that without looking at the distribution first.
3. Why risk does not follow the calendar
A date range is a poor main strategy for risk, because risky items are not spread evenly across years.
Flagged items cluster in ways that have nothing to do with the calendar. Someone may have had a year of moving house and posting about it, or a stretch where a job made them vent publicly. Run a check before setting a cut, and look at how flagged items distribute across years rather than relying on memory.
Watch the boundary of the range as well. Tools differ on whether posts from the cut-off day itself are included, so check the estimated count before committing.
4. Three checks to run before you start
Three checks before starting: confirm the estimated count, confirm the archive opens, and confirm the job can be stopped and resumed.
The estimated count is the fastest sanity check available and it takes seconds. If the tool says this run will touch forty posts and your instinct says four hundred, the range is wrong. Find that out before the job starts, not after.
Confirm the archive opens cleanly and the files inside are complete, and keep it on local storage rather than a synced folder. Then check whether the job supports pause and resume. The smaller the scope, the easier it is to misjudge, and a stoppable job is what covers that.
5. Combining two scopes without losing track
Date-based and risk-based scopes combine well as batches, and mixing both inside a single run is where people get into trouble.
A workable pattern is two or three date-based batches inside a risk-based category. Take the identity items first regardless of year, split them into batches, run the earliest batch and watch how the job behaves before continuing.
If the goal is a lower score, risk-based selection handles it directly. If the goal is a timeline that reads as clean, date-based is more efficient. Pursue both if you like, but do not set both kinds of selection in the same run, because you will not be able to tell which criterion did what.
Practical takeaways
- Scope decides both the outcome and how reversible the cleanup is, which makes it a more consequential decision than which tool you use.
- Deleting everything fits leaving a platform, deleting by date fits traces tied to a period, and deleting by risk has the best hit rate while depending on a report you trust.
- A date range is a poor main strategy for risk, because risky items are not spread evenly across years.
- Three checks before starting
- Date-based and risk-based scopes combine well as batches, and mixing both inside a single run is where people get into trouble.
Whichever route you pick, run the estimated count first. It is the cheapest mistake to catch in the entire process, and it is the one people skip because the range looks obviously right.
The longer version with the reference detail is here: https://digital-footprint-health.shop/blog/deletion-scope-selection
