Did marketing email open tracking suddenly change in HubSpot?

Marketing & Content Hub Email Deliverability
1

Hi HubSpot Community,

I’m trying to determine whether there was a change to HubSpot’s marketing email tracking, bot filtering or delivery behaviour in early June 2026.

Until early June, our marketing emails consistently recorded open rates of approximately 16–20% excluding bots. We did not intentionally change our sender, sending domain, list, email process or email format.

However, between our 5 June and 12 June sends, the reported open rate dropped abruptly and has remained around 3–4% ever since.

5 June 2026
Sent: 181
Delivered: 180
Open rate excluding bots: 16.67%
Unique opens excluding bots: 30
Open rate including bots: 30.56%

12 June 2026
Sent: 180
Delivered: 179
Open rate excluding bots: 3.91%
Unique opens excluding bots: 7
Open rate including bots: 12.85%

19 June 2026
Delivered: 180
Open rate excluding bots: 3.33%
Unique opens excluding bots: 6
Open rate including bots: 12.22%

26 June 2026
Delivered: 180
Open rate excluding bots: 4.44%
Unique opens excluding bots: 8
Open rate including bots: 15.00%

28 August 2026
Delivered: 182
Open rate excluding bots: 3.30%
Unique opens excluding bots: 6
Open rate including bots: 13.74%

The important point is that this was not a gradual decline. It was a clear step-change between 5 and 12 June and has persisted ever since.

I have checked the following:

  • Bot filtering is ON
  • Bot filtering was also ON on the 5 June email when the open rate was still 16.67%
  • Data Privacy settings are OFF
  • Delivery volume has remained essentially unchanged
  • There was no intentional change to our sender, sending domain or normal sending process
  • Sender: Present WAY Coaching
  • From address: [email]
  • Domain: presentwaycoaching
  • Location: Australia

I have also recently started seeing the warning during Review and Send:

“Check whether email open or click tracking requires prior recipient consent for your recipients. If required consent has not been obtained, disable tracking before sending.”

I understand this may simply be a compliance notice, as Data Privacy is OFF, but I’m including it in case it relates to recent changes in email tracking.

I would particularly like to understand:

  1. Did HubSpot change its marketing email open-tracking methodology around late May or early June 2026?
  2. Did HubSpot change its bot-detection or bot-classification logic during this period?
  3. Did the treatment of Apple Mail Privacy Protection, proxy opens, security scanners or automated opens change?
  4. Are some events now removed before they reach the “including bots” figure?
  5. Could a change in HubSpot’s shared sending infrastructure, tracking domain or sender reputation explain the sudden reduction?
  6. Has anyone else seen a similar abrupt change around this date?

The part I find particularly difficult to explain is that even the including-bots open rate fell from 30.56% to approximately 12–15%, while deliveries remained almost identical.

Any insight from HubSpot staff or other users experiencing the same thing would be greatly appreciated.

Thanks,
Justin
Present Way Coaching

Moderator note: For information, the HubSpot Community is a public space. I have therefore edited your content for security & confidential reasons.

2

Hey Justin - you have probably already checked most of this. Asking anyway, because a cliff that steep with flat delivery is usually a variable that moved, not content overnight.

Basics first. Same audience definition as the pre-cliff sends (list, suppression, segments)? Same from address and brand domain? Domain health / reputation (Gmail Postmaster, Microsoft SNDS) still clean? Spam-folder / blocklist checks clear on both sides of June? If any of those changed between 5 and 12 June, start there.

If those are exhausted and unchanged:

  1. Compare click rate and reply/meeting outcomes on the same sends. Clicks held, opens cratered → treat opens as a broken proxy.
  2. Side-by-side one pre-cliff and one post-cliff email: tracking pixel, template type, brand vs HubSpot tracking domain, list definition.
  3. Note when the Review-and-Send consent/tracking warning first appeared. Breadcrumb, not proof opens were suppressed.
  4. Ask Support for open + bot-classification history around 5–12 June with those two email IDs. Staff can see accounting/infra changes; Community usually cannot.

One fact splits the cases: did clicks fall with opens, or stay flat? If you can paste those two performance tables, happy to sanity-check.

Moderator note: While this solution may not address the original poster’s specific situation, it could be helpful for other community members facing similar challenges.

4

Hi @JustinDwyer78 and welcome, we are so glad to have you here!

Thanks for reaching out to the HubSpot Community! And thanks @Derek_Reynolds for your help!

I’d like to share a few points that may help separate a tracking change from an inbox-placement problem:

  • HubSpot counts an open when the email images load, or when a recipient clicks a link without the images loading. If the recipient’s mail client, privacy service, or corporate security layer does not load the tracking pixel, the open will not be recorded. See Analyze marketing email recipients.

  • With bot filtering enabled, HubSpot excludes suspected non-human activity. Filtering affects the engagement data shown in reports. See Understand bot filtering in marketing email analytics.

  • The consent warning is separate from bot filtering. It is a compliance prompt about whether tracking is lawful for the recipients; with Data Privacy settings disabled, it would not by itself explain a sudden change between two sends. It is still important to review the consent requirements that apply to your audience and location.

Because your including-bots rate also fell, comparing the 5 June and 12 June emails across the following would be helpful:

  • click rate and the individual recipients who clicked;

  • open rate by email client/device, where available;

  • engagement over time, especially opens immediately after delivery;

  • bounce, spam-report, unsubscribe, and deferred-delivery data by recipient domain; and

  • a few known recipients who can confirm whether the message arrived in the inbox and whether images were loaded.

If clicks, replies, conversions, and inbox-placement signals stayed stable while both open views fell, that would point more toward a tracking or email-client measurement change. If clicks and other engagement also fell, particularly for one mailbox provider, then sender reputation or inbox placement should be investigated as well. Also check the message size and HTML, authentication records, sending-domain status, and whether the tracking domain changed.

I’m checking whether any tracking or bot-classification updates occurred during that window, between 5 and 12 June 2026 and I’ll get back to you if I have additional information.

In the meantime, I would use clicks, replies, conversions, and website activity as the primary engagement signals rather than treating open rate alone as a definitive measure of readership.

I’d love to hear from our Top Experts: Hi @danmoyle, @Josh and @BarryGrennan do you have any insights to share, please?

Thanks so much and have a lovely day!

Bérangère

5

Thanks for the tag @Berangere but honestly, nothing to add. I consult the same way you laid it out. Opens have become for too unreliable for me and I encourage clicks and website activity (and replies where applicable) as better metrics.