Interpreting "𝗥𝗲𝗮𝗱, 𝗦𝗸𝗶𝗺𝗺𝗲𝗱, 𝗚𝗹𝗮𝗻𝗰𝗲𝗱" in your performance report📊

[Roynal]

Performance reports are fun to dissect in email marketing. HubSpot provides an interesting attention metric called Time spent viewing email (Read, Skimmed, Glanced).

Here's my interpretation of this metric:

• Read (8+ seconds) - > More of this means my content is interesting❗❗. However, I add one more consideration, the click rate↗️ of the email. If it's too little then my content is not actionable.
• Skimmed ( 2-8 seconds) -> My subject lines are probably good because it is driving attention⚠️, but content-  not so much👎 and could be improved.
• Glanced (0-2 seconds) -> Not good anything☠️

One more thing to be aware of is that your audience is always going to be spread between all 3 classes of this metric. My goal is to have more of my audience in the "Read and Skimmed" buckets as much as possible👍.

What is your interpretation?

[SLane1]

I'm also wondering why this metric's subset counts don't add up to the total count of opens. For example, I have an email that was opened by 31 people. I see 16 read it and 2 skimmed and 0 glanced. Why does the sum of the subset count not add to 31, with some opens unaccounted for?

[DCurtis8]

I have the same issue! Does anyone know why this happens? There are some contacts I want to manually re-engage with who have opened my emails, but I can't see how interested they are in the content.

[SLane1]

Does anyone know if it's possible to export the read, skimmed, glanced data in mass for all emails as a .csv? I can't seem to access the attention metric in any reporting tools and don't see a mass performance r. 

[Sugumar_Renu]

Hello @Roynal 

Just curious, How we calculate the read time in mail clients. Is this based on clicks in the email ? 

[Atomiciser]

I'd like to bump this to the top again.  How is this calculated?  We don't need to know the algorithms involved but would be good to have an idea to reassure me that it's accurate.

As far as I know:

• Every mainstream email clients will block any tracking code in emails as it's a security violation.
• Therefore it's impossible to track user behaviour using any kind of script: even if the script could run, it would be blocked from returning that data to the server due to security.  So we can't see how long the email was open for, how far they scrolled or what they hovered over etc.
• We CAN track image downloads on email because they initiate a web request which can be measured by server.  SOME email clients (mostly phone apps) will lazy load images and only load an image when the user scrolls to it.
• If links are clicked, these can show some engagement too, such as how far down the page the user has reached etc.
• The read, skimmed, glanced tags can't be generated off clicks, because emails are still tagged even if they don't get any clicks.

Therefore, I would interpret the "Read, Skimmed, Glanced" differently:

• Read: most or all the images in the email have been downloaded, with a time gap between them, indicating the email client "lazy loaded" the images as the user slowly scrolled down the email.
• Skimmed: The user quickly scrolled down the page, prompting all images in the email to load very quickly / all at once.
• Glanced: The first few images on the email (above the fold) have been loaded, but none others.

HOWEVER, the "Skimmed" version will generate false positives, because many desktop email clients will download all images at once when the email is first opened.  Therefore an interested reader using Outlook on their laptop will be marked as "Skimmed" even if they read the email through thoroughly.  So maybe we should take this tag with a pinch of salt and not a reliable metric?

I'd be very interested to know others thoughts on this subject.

[Jnix284]

@Atomiciser I think your definitions align more closely to the technical assumptions / algorithms that determine the metrics, whereas @Roynal was generalizing what the implications are to guide improvements on future emails.

I haven't found anything concrete as to how they are determined and have always viewed them as a nice-to-have insight. As they aren't KPIs, I've never included them in any performance reporting and typically used them as supporting evidence when reviewing an email that didn't perform as expected or was an outlier.

While all browsers / email clients don't load images the same way, it would be my assumption that this is also accounted for in the algorithm since the browser/client can be detected.

Definitely grain of salt data, with the recent changes I wouldn't be surprised if this data disappears.

[danmoyle]

I like your interpretation @Roynal. Seems logical. I especially like the addition of clicked links. That's what it comes down to for me. Each email should have a goal. Is it a reply? A click? A meeting booked? If we set this expecation ahead of time, we know each email ought to end in a click at some point. With the evolution of privacy settings, opens and glances and reads are becoming less trustworthy. Those clicks though... hopefully that will help with insights. 

Love the conversation! 

[Jnix284]

@Roynal I think your interpretations are pretty spot on - the only thing I would add/change is the following:

Glanced: they want to hear from me, but this message didn't hit the target

I'll take Glanced over not opened, it is still telling me something about that contact that distinguishes them from those who didn't even open.

[Roynal]

Thanks for the note, and yes I agree on Glanced :).

[kvlschaefer]

Thanks for sharing, @Roynal!

I'll add some top contributors from our community to get the conversation going. 

Hi @Jnix284, @Ben_M, @louischausse, @danmoyle, @GeorgeBThomas, @myles4, @Ian_Matt, @Josh - How do you interpret the email performance report? Do you have any thoughts for @Roynal?

Thanks!

Best,

Kristen