Your LINE friend acquisition route report is showing “other” for almost everything because LINE suppresses any route that received fewer than 20 additions in the period you selected, and it applies that threshold three separate times over. Anything suppressed does not disappear from the totals. It gets swept into “other”. So the more carefully you tagged every poster and every ad, the more thoroughly you broke the report, because you split one visible number into a dozen invisible ones.
There are four other rules underneath that one, and each of them can independently produce the same symptom. This is the setup guide written backwards from the failure, so that you configure routes on a LINE Official Account in a way that survives contact with LINE’s privacy floor rather than being erased by it.
Table of Contents
The 20-addition threshold, applied three times over
LINE’s own Japanese manual states the rule for routes you create yourself under the friend add guide. Across the whole period you selected, each of route, traffic source and campaign must have recorded 20 or more additions or blocks before any number is shown. Below 20, the number is not shown, and the route is counted into “other” while the traffic source and campaign are counted into “other” as well.
Then there is a second, stricter version of the same rule for the daily line graph. For each individual day, that day needs 20 or more additions or blocks before it appears. Below 20, that day is drawn as zero.
| Where the threshold applies | What it needs | What happens below the line |
|---|---|---|
| Route, across the selected period | 20 or more additions or blocks | Reclassified into “other” |
| Traffic source, across the selected period | 20 or more | Reclassified into “other” |
| Campaign, across the selected period | 20 or more | Reclassified into “other” |
| Any single day, on the line graph | 20 or more that day | Drawn as zero for that day |
| Novelty goods and poster routes, sub-route detail | 20 or more per traffic source and per campaign | Sub-route detail unavailable |
LINE’s own worked example of the problem
The manual illustrates this with a homepage route, and the illustration is more useful than any explanation. Over the selected period the homepage route brought 50 friends. On 23 May it brought 40, which is above 20, so 23 May shows 40 on the line graph. The other 10 additions happened across various other days, none of which reached 20 on its own. Those 10 are drawn as zero.
The pie chart and the table still show 50. The line graph shows 40. Both figures are correct. Read only the daily graph and you will conclude the route died on 24 May.
Why adding more tags makes the report worse
Every instinct from Google Analytics says tag granularly. Give each poster its own parameter, each ad set its own campaign name, each store its own code, and sort it out later in the data.
That instinct is exactly wrong here, and it is the single most common way operators destroy their own report. The threshold is applied per tag. Ten in-store posters at 15 additions each over 90 days gives you 150 friends and zero visible rows, all 150 dumped into “other”. One poster route with ten traffic sources beneath it, each carrying 15, gives you a visible route of 150 and invisible sources underneath. One route with two traffic sources at 75 each gives you a complete picture.
So consolidate before you tag. Work out the smallest number of buckets that still answers a decision you would actually make, and build routes at that level. If splitting Taipei from Kaohsiung would not change your spend, do not split them.
There is a second reason to keep buckets fat. Segmented message delivery needs a target reach of 100 or more, and it will not send if the filtered audience falls below 50. A route too small to display is also a route too small to message.
Nothing you tag today applies to friends you already have
Route tagging is not retroactive. Creating a new route, or a new parameterised add-friend URL, does not relabel anybody who is already a friend. Only people who come through that specific URL or QR code after it exists get attributed to it.
This is the reason so many operators tag everything correctly and still see nothing change. The report you are staring at is describing acquisitions that happened before your tags existed, and it will keep describing them until enough new tagged additions accumulate to move the mix. On a 90-day window, that means waiting most of a quarter before the report reflects your setup.
The dates before which no route data exists at all
There is a hard historical floor. Route analysis only covers friends added on or after 15 March 2020. Anybody who joined before that has no route and never will.
Individual routes then have their own later start dates, published in the manual, and additions before those dates sit in “other” permanently.
| Route | Counted from |
|---|---|
| Coupons | 28 February 2024 |
| Shop card | 12 June 2024 |
| Yahoo! Shopping | 10 July 2024 |
| Add-friend URL, QR code and button, including profile link copy | 13 November 2024 |
| Recommended official accounts inside the profile | 20 November 2024 |
One more measurement condition rarely gets mentioned. Route measurement requires LINE app version 10.0.0 or later on the user’s phone. Somebody running an older build who adds your account still counts as a friend, and their route is simply never recorded.
The 90-day rule that quietly poisons the breakdown
This is the one that will make you doubt the report even after you have fixed everything else. If a user has been a friend for 90 days or more and then blocks you, that block is counted into “other” rather than against the route that originally acquired them.
The consequence is documented by LINE as a frequently asked question, because it produces a result that looks impossible. The “other” row can show more blocks than additions. Nothing has broken. What you are looking at is long-tenured churn from every route in your account, pooled into a single bucket where you can no longer tell which channel any of it came from.
Practically, this means your route breakdown gets less trustworthy as the account gets older. A three-month-old account has clean attribution. A three-year-old account has a permanent stream of mature churn flowing into “other”, inflating that bucket in a way that has nothing to do with your tagging.
There is a related trap in the same screen. The block count shown is simply the number of blocks that occurred during the selected period. Those people could have joined at any time, possibly years earlier, so the figure has no relationship to the additions sitting next to it. You cannot read a cohort block rate off this report, so please do not calculate one from it.
How far back each screen reaches, and how stale it is
| Screen | Window | Freshness |
|---|---|---|
| Friends overview, additions, target reach and blocks | Up to 397 days selectable | Most recent 7 days shown by default |
| Target reach attributes, gender, age and region | Same | Roughly 3 days behind |
| Friend acquisition route | Up to 90 days selectable | Approximate when filtered, see below |
| Friend add ads performance | Past 5 years only | May disagree with the route screen depending on the user’s app version |
Two of those rows change how you should work. The attribute breakdown of target reach is roughly three days behind, so nothing you did this week is in it yet, and judging a campaign that launched on Monday by Wednesday’s demographics is judging last week. And when you apply a filter to the route screen, the numbers are displayed as approximations rounded to units of 50, so treat a filtered view as directional and never paste it into a report.
The 90-day ceiling on route analysis against a 397-day ceiling on the overview also means the two screens can never be reconciled across a full year. Download the route data monthly if you want a longer series. The manual confirms both the summary and the daily data are downloadable, which is the only way to build a trend beyond a quarter.
What is genuinely inside “other”
Some of it is legitimate and unfixable. LINE describes “other” as the total of friends acquired through its smaller services and through non-primary paths in the app, giving LINE Beacon and additions from non-mobile devices as examples, and states outright that it does not disclose every route that falls into the bucket.
So a residual “other” is normal and always will be. What is not normal is “other” dominating the chart, and when it does, the cause is almost always one of the five rules above and almost never a mysterious channel.
What to change, in order
- Count the routes you have created. If any of them are receiving fewer than 20 additions per 90-day window, merge them until they clear the threshold, and push the detail down into traffic source and campaign only where those will clear 20 as well.
- Set your reporting window to the full 90 days rather than the last week, because the same data crosses the threshold at 90 days and fails it at 7.
- Accept that the next 90 days is your first real measurement period, since nothing you tag now applies backwards.
- Stop reading block numbers as a quality signal per route once your account is over a year old, because mature churn is being pooled into “other” regardless of source.
- Export the summary and daily data every month, so that the 90-day ceiling stops being the limit of what you know.
