Instagram report · @bluebottle · October 2026
Blue Bottle published 5 feed posts in October against 12 in September, yet the median post earned 974 likes versus 569.5 while total likes fell 75.5% with fewer posts published.
Reporting period 1–31 October 2026 · comparison period 1–30 September 2026 · public data collected 6 November 2026, 17:32 UTC from Instagram through Nolyn. 17 posts analysed across the two periods; 1 post excluded. Figures are public likes, comments and Reel views only.
Coverage and method
Both periods are fully covered: the post list runs past 1 September, so no figure depends on missing posts. One September collaboration post, owned by another account, is excluded.
Executive summary
Blue Bottle published 5 posts in October — 1 photo, 3 carousels and 1 Reel — against 12 in September, so total likes fell 75.5% to 4,322 and total comments fell 52.0% to 71. On a per-post basis the account did better, not worse: median likes rose 71.0% to 974, median comments rose 50.0% to 12 and median engagement per post rose from 0.111% to 0.193% of the 519,225 followers. The best post was the 22 October carousel pairing the Giant Steps blend with the Notes in Repose jazz album, at 1,158 likes and 10 comments — 0.225% engagement — and the weakest was the 28 October carousel introducing Guatemala Antigua Vandaluz with 420 likes and 12 comments. The one thing to do next is publish more often without losing that quality: November should carry 10–12 posts, weighted toward the carousel format that produced October's highest likes.
Follower counts from past periods are not public, so engagement in both periods is measured against today's 519,225 followers and no follower growth is reported. October's last post went live on 30 October, so it has had less time to collect likes than the September posts did — October's per-post advantage is if anything understated, but the comparison of recent posts is never exactly like-for-like.
Scorecard
Counts of posts are shown in posts, because they sit under 10. All other changes are percentages. Each median notes how many posts it rests on.
| Metric | Oct 2026 | Sep 2026 | Change |
|---|---|---|---|
| Posts publishedn = 5 vs n = 12 | 5 | 12 | −7 posts |
| Photosn = 1 vs n = 3 | 1 | 3 | −2 posts |
| Carouselsn = 3 vs n = 5 | 3 | 5 | −2 posts |
| Reelsn = 1 vs n = 4 | 1 | 4 | −3 posts |
| Total likes | 4,322 | 17,637 | −75.5% |
| Total comments | 71 | 148 | −52.0% |
| Median likes per postn = 5 vs n = 12 | 974 | 569.5 | +71.0% |
| Median comments per postn = 5 vs n = 12 | 12 | 8 | +50.0% |
| Median engagement per post(likes + comments) ÷ 519,225 followers · n = 5 vs n = 12 | 0.193% | 0.111% | +73.6% |
| Median Reel viewsn = 1 in October — too few to judge | 45,405 | 28,613 | not shown1 Reel only |
Source: Instagram public post data for @bluebottle, collected 6 November 2026, 17:32 UTC. Engagement uses today's follower count in both periods. October's median Reel views rest on a single Reel, so no change is reported.
Read the two halves of this table together. Every total fell, but totals fall automatically when you publish 7 fewer posts. Every median rose, which is the more meaningful signal for a brand account of this size — it says a typical October post drew more likes and comments per follower than a typical September post, even though September contained the largest post of either month (a 16 September Reel at 9,110 likes and 132,973 views) that pulls September's total up sharply.
Bars scaled to 1,158 likes; format shown beside each value.
By format
Every format rests on fewer than 5 posts, so none is enough to judge: three carousels, one photo and one Reel.
| Format | Posts | Median likes | Median comments | Median engagement | Median Reel views |
|---|---|---|---|---|---|
| Carouseltoo few to judge | 3 | 974 | 12 | 0.193% | — |
| Phototoo few to judge | 1 | 636 | 17 | 0.126% | — |
| Reeltoo few to judge | 1 | 1,134 | 6 | 0.220% | 45,405 |
| All October posts | 5 | 974 | 12 | 0.193% | 45,405 |
Engagement = (likes + comments) ÷ 519,225 followers, median of the posts in each format. Reel views exist only for the Reel format. Fewer than 5 posts in a format is marked too few to judge.
What was published
Every post is listed newest first with its public likes, comments and, for the Reel, views. Captions are summarised, not quoted in full; carousel slides were not individually available.
| Date | Format | What it is about | Likes | Comments | Views | Link |
|---|---|---|---|---|---|---|
| 30 Oct 2026 | Reel | Annual Holiday Collection teaser: limited-release coffees, merch, curated gifts; subscribe to shop first | 1,134 | 6 | 45,405 | DfIfM-JS4EZ |
| 28 Oct 2026 | Carousel | New Signature Origin Guatemala Antigua Vandaluz: farm story, tasting notes, in cafes or whole bean | 420 | 12 | — | DfDbKIpAb2y |
| 22 Oct 2026 | Carousel | Giant Steps blend paired with the new Notes in Repose jazz compilation album | 1,158 | 10 | — | Dez9jLxAThE |
| 20 Oct 2026 | Photo | Matcha Essentials Kit: Uji ceremonial matcha, whisk, bowl, tea towel | 636 | 17 | — | DeuzvYKkhgP |
| 6 Oct 2026 | Carousel | Notes in Repose vinyl release, “out now”: eleven tracks from nine cities, limited edition | 974 | 26 | — | DeKdNpcD-XS |
| 5 posts | 1 photo · 3 carousels · 1 Reel | Totals for the reporting period | 4,322 | 71 | 45,405 | — |
Source: Instagram public post data for @bluebottle, collected 6 November 2026, 17:32 UTC. Views are shown for Reels only; they are not reach. Publish dates are UTC calendar days as reported by the platform.
Best and weakest
Ranked by engagement, (likes + comments) ÷ 519,225 followers. With only five posts the middle one is both third-best and third-weakest; the overlap is shown rather than hidden.
The caption puts a named coffee next to a named album and a suggested moment — a morning with nowhere to be — instead of describing the coffee on its own.
Why it matters: it is the only October post whose caption pairs two of the brand's own products, and it drew the most likes of the month at a normal comment level for this account.A four-second Reel with a scarcity-and-subscribe call to action, pointing at a collection that had not yet launched.
Why it matters: highest views of the month, but comments are low for its likes — see the flag below before treating it as proof that Reels work.A launch announcement with the strongest comment response of the month (26), listing the physical details of the record and pointing to the link in bio.
Why it matters: the most commented post of October, which is the metric that moved least across both periods.A long origin story naming the producers and the volcano, with tasting notes and a buy path — the most information-dense caption of the month.
Why it matters: the lowest likes of the month at the month's median comment count. The data cannot say why.A single still with a list-style caption itemising the four pieces in the kit.
Why it matters: the only photo of the month. One post cannot establish that photos underperform.The same post as Best 3: with five posts, the middle of the ranking falls in both groups.
Why it matters: nothing here separates it from the top; the third-best and third-weakest label is an artefact of a five-post month.Flags and caveats
Next month
Every baseline below comes from a median or a count across ordinary posts, never from the flagged 30 October Reel. No results are forecast.
October's per-post medians were strong but the month shipped 7 fewer posts than September, which is why total likes fell 75.5% to 4,322. September showed the account can sustain 12 posts and still produce a median of 569.5 likes; October's median of 974 came from fewer than half as many posts. Return to roughly 2–3 posts a week.
Check in the November report: posts published (target 10–12) and total likes (October baseline 4,322; September 17,637).Carousels were 3 of October's 5 posts and carried a median of 974 likes and 12 comments; the account's best post of the month paired the Giant Steps blend with the Notes in Repose album rather than describing one product. Aim for at least three carousels in November, each pairing two of the brand's own products or a product with a cultural object.
Check in the November report: carousel count of 3 or more, and carousel median likes at or above 974 with median comments at or above 12.October contained a single Reel, which is too few to judge, and that one carries a possible-promotion flag, so it cannot be used as evidence. September's four Reels gave a median of 28,613 views — that is the honest baseline for the format on this account.
Check in the November report: number of Reels (target 3 or more) and median Reel views against 28,613, with the same flag test applied to any outlier.Ready to send
Copy the text below into a mail to the client; the figures match the tables above.
Hi [name],
Here is the October Instagram report for @bluebottle, covering 1–31 October 2026 against 1–30 September 2026. We read every post in both periods: 5 in October and 12 in September, plus one September collaboration post owned by Nespresso that is excluded from the figures.
October was a quieter month — 5 posts, 1 photo, 3 carousels and 1 Reel — so total likes came in at 4,322, down 75.5% on September's 17,637. Per post, the account did better than in September: median likes rose 71.0% to 974, median comments rose 50.0% to 12, and median engagement rose from 0.111% to 0.193% of the 519,225 followers. The strongest post was the 22 October carousel pairing Giant Steps with our Notes in Repose album (1,158 likes, 10 comments); the weakest was the 28 October Guatemala Antigua Vandaluz carousel (420 likes, 12 comments).
Two things to keep in mind when reading the numbers: engagement in both periods is measured against today's follower count, because past follower counts are not public, so there is no follower-growth figure here; and the 30 October Reel collected views well above September's median with very few comments, which may reflect paid distribution we cannot see publicly. It is included in the totals but not used to draw conclusions about format.
For November we suggest 10–12 posts, at least three carousels built on two-subject captions, and at least three Reels measured against September's median of 28,613 views. Each recommendation in the report states the number we will check in next month's edition.
Happy to walk through any of it.
[sender]
Sources and limits