Seventy six of 170 title records sit in a cluster, but only 23 screen records sit beside another screen record.
Only six clusters hold more than one screen record. So 23 of 144 screen records, 24 percent, sit beside another screen record. That is the franchise figure, and it is less than half the headline 54 percent.
Seventy six of the 170 title records on this site sit in a multi record cluster, and 56 stand alone. But only 23 screen records share a cluster with another screen record. Most of what looks like franchise structure is one drama filed next to the book it came from.
The question here is narrow: how much of the catalogue connects to something else, and what kind of connection it is. A sequel chain, a set of parallel versions of one source, and a drama filed beside its own novel are three different structures, and the catalogue holds far more of the third than the first.
Every title record carries a related_content field listing other records on the site. Treating each entry as an undirected edge and finding the connected groups across all 170 records gives 78 groups. Fifty six of those are a single record with no link at any distance. The remaining 27 groups hold 76 records between them.
| Cluster size | Clusters | Records |
|---|---|---|
| 1 (stands alone) | 56 | 56 |
| 2 | 16 | 32 |
| 3 | 3 | 9 |
| 5 | 1 | 5 |
| 7 | 1 | 7 |
| 12 | 1 | 12 |
| All clusters of 2 or more | 27 | 76 |
State the exclusions plainly. This rule reads one field and nothing else. Records sharing a studio, a cast member, a director or a performer pairing are not clustered. Two adaptations of books by the same author are not clustered unless the field says so. A second season that was never given its own record cannot be clustered, because there is nothing to link to.
Split the 170 records by whether they are a screen record or a printed source record. The catalogue holds 26 source records: 12 web novels, 7 novels, 5 manga, 1 manhwa and 1 manhua. All 26 are in a cluster, every one of them. Of the 144 screen records, 39 are.
That asymmetry is the whole story. A source text gets a record here only when the site also holds an adaptation of it, so a source record is clustered by construction. Sixteen of the 27 clusters are exactly this: one screen record plus the book, comic or web novel it adapts, 33 records in total. GAP and GAP (Novel) is the shape. So is Payback, which reaches three records because both a manhwa and a novel sit behind it, mapped in our Payback reading guide.
Only six clusters hold more than one screen record. They account for 32 records, 23 of them screen records. So 23 of 144 screen records, 24 percent, sit beside another screen record. That is the franchise figure, and it is less than half the headline 54 percent.
| Cluster | Records | Screen | Source | Relation between the screen records |
|---|---|---|---|---|
| Fourever You | 12 | 7 | 5 | Numbered continuation plus arcs and specials |
| Cherry Magic | 7 | 6 | 1 | Parallel versions of one manga, plus a film sequel |
| Mo Dao Zu Shi | 5 | 3 | 2 | Parallel versions of one web novel, plus a spin off |
| Love Class | 3 | 3 | 0 | Three numbered seasons |
| Unexpectedly Naughty Fukami | 3 | 2 | 1 | Anime and live action from one manga |
| Only Boo! and WEIRDO-101 | 2 | 2 | 0 | Linked productions with separate source novels |
Two of the six carry a numbered continuation: Fourever You Part 2 follows Part 1, and Love Class Season 2 and Love Class Season 3 follow the first. Three are parallel versions of a single source rather than a story moving forward. The Fourever You set is the largest at twelve records and has its own watch order guide.
A continuation can be recorded two ways: as a second season counted on one record, or as a second record. The catalogue does both, and the counts do not line up. The seasons field holds “1” on 86 records, “2” on four, “3” on one, and is blank on 30. The 30 blanks are all formats that do not have seasons: the 26 source records plus 2 movies, an anime movie and a musical. So only five records in the catalogue report more than one season internally.
Three of those five stand alone in the cluster graph: Blank, Fragrance of the First Flower and GELBOYS each report two seasons inside a single record, so their continuation is invisible to any rule built on links. The reverse also happens. All three Love Class records report seasons as “1”, because each record describes its own season, so a rule built on the seasons field would miss that chain entirely. Thirty four clustered records report exactly one season.
Neither method gives a complete count. Combining them, at least seven distinct stories here have more than one installment recorded. That number is a floor.
Counting only the first country a record lists, here is the cross tabulation, alongside whether the record’s adaptation_source field names a prior work it is based on.
| Group | In a cluster | Stands alone | Total | Clustered |
|---|---|---|---|---|
| Thailand | 38 | 31 | 69 | 55% |
| Japan | 17 | 2 | 19 | 89% |
| China | 5 | 6 | 11 | 45% |
| South Korea | 5 | 8 | 13 | 38% |
| Taiwan, Hong Kong, Vietnam, Philippines | 0 | 9 | 9 | 0% |
| Screen records naming a prior source | 35 | 21 | 56 | 62% |
| Screen records naming no prior source | 4 | 34 | 38 | 11% |
The adaptation split is the sharpest cut in the data, and it needs a caveat immediately. A screen adaptation is clustered when the site also holds a record for its source, and 21 of the 56 adapted screen records have no such sibling. The field measures cataloguing depth as much as production structure. The four clustered screen records naming no prior source are the three Love Class seasons and a behind the scenes special.
Japan’s 17 of 19 is the highest rate, and it is concentrated: 6 of those 17 are Cherry Magic records, including the source manga. Take the Cherry Magic cluster out and Japanese screen records fall to 6 of 8, still high but on a sample too small to lean on.
First, the expectation that franchise structure would concentrate in the country with the most records. It does not. Thailand holds 69 of 170 records under the first country rule, 57 percent of the catalogue, and 38 of the 76 clustered records, 58 percent of them. Its clustering rate of 55 percent is one point off the catalogue average of 54 percent. On that measure Thailand is exactly average, and the hypothesis fails.
Second, the expectation that records inside a cluster would be carried on more platforms. The median platform_count is 2 for clustered screen records and 2 for standalone ones, which is also the catalogue median. The means differ, 2.54 against 1.82, but that gap survives largely because the Cherry Magic and Mo Dao Zu Shi records carry long platform lists. On the median test the groups are indistinguishable, so this is not reported as a finding.
A third check was too small to settle. Among screen records, 6 of 12 GL records are clustered against 31 of 77 BL records. The direction is there, the sample is not.
Sample: all 170 title records on blglflix.com, read on 15 September 2026. Fields read were related_content, seasons, title, original_title, content_type_text, country_text, adaptation_source and platform_count, plus the computed franchise_titles seam, which matched related_content on all 170 records with no discrepancy.
Four limitations. The graph has one asymmetric edge: Only Boo! lists WEIRDO-101 in its related content, and the WEIRDO-101 record lists nothing. Reading edges as undirected forms the pair anyway, but a directed reading would leave both standing alone, which would change the 65 to 63.
Second, the adaptation split relies on reading prose. The adaptation_source field is a paragraph, not a flag, so records were classified by searching for phrases such as “adapted from” and “based on” after stripping negated forms such as “not based on”. One record, You Are My Present, has the field empty and falls in neither group.
Third, the result is time skewed. Records with a first release year from 2020 to 2023 are clustered 25 times against 4 standalone, while 2026 records are clustered 15 times against 39 standalone. Newer titles have had less time to acquire a sequel and less time for their source records to be added, so 54 percent is a snapshot of cataloguing progress as much as of the genre.
Fourth, the seasons disagreement above means no single field answers the question “how many continuing stories are here”. At least seven is the honest form of that answer.
adaptation_source fields on all 170 records for the adaptation classification and for the relation described in each cluster.
Seventy six of the 170 title records are linked to at least one other record, and 56 stand alone. But 33 of those 76 sit in a cluster that is just one screen title plus its printed source, so the count of records that sit beside another screen record is 23.
Fourever You, at 12 records: 7 screen records and 5 web novels. Cherry Magic is second at 7 records, and the Mo Dao Zu Shi set, which includes The Untamed, is third at 5.
Each record’s related_content field was read as an undirected edge to another record, and the connected groups were found across all 170 records. That gives 78 groups, 56 of them a single record. No other signal was used.
No. Shared studio, shared cast, shared director and shared performer pairing are all excluded. Only the explicit related_content links count.
Because a continuation can be filed either way. Five records report more than one season inside a single record, and three of those five have no cluster links at all. Meanwhile all three Love Class records report one season each while forming a three record chain.
Among the 144 screen records, 35 of the 56 that name a prior source are in a cluster against 4 of the 38 that name none. The caveat is that an adaptation clusters only when the site also holds a record for its source, so the figure partly measures how deeply the catalogue is filled in.
No. Thailand holds 57 percent of the catalogue and 58 percent of clustered records, and its clustering rate of 55 percent is one point off the catalogue average. That check did not hold up.
It does not rank or judge any production, and it makes no claim about how many sequels exist in the wider BL and GL world. Every figure describes the 170 records held on this site on 15 September 2026.
Metadata and images in this article come from TMDB and official sources, and remain the property of their copyright holders. This product uses the TMDB API but is not endorsed or certified by TMDB. BLGLFLIX is an independent site with no affiliation to the productions covered. Full copyright, correction and takedown policy: Disclaimer.
Discussion 0 comments
Join other fans talking about Franchise and Sequel Structure: How the Records Cluster. Be kind, no spoilers without tags, no spam.
No comments yet.
Share the first reaction below.