Your tracker just pinged. Someone opened your email. Before you fire off that follow-up, it's worth asking: was it actually a person?
In 2026, a lot of opens aren't. Raw open rates on B2B lists run 2-3x inflated because machines load [tracking pixels](/glossary/email-tracking-pixel) too. The good news is that machine opens leave fingerprints. Here's how to spot them, step by step.
Step 1: How fast did the open happen?
Check the gap between your send time and the open. Seconds? That's a machine.
Corporate security scanners like Mimecast, Proofpoint, and Microsoft Defender process every incoming email on arrival, opening every link and image to check for malware. Between 15 and 25 percent of B2B recipients sit behind one. Their "opens" land almost instantly, every time, on every email.
Humans are slower. They open minutes or hours after you send, at hours when people are actually awake. An open at 2:47 p.m. on a Tuesday is plausible. An open four seconds after send is not.
Step 2: Did every link get hit at once?
Look at what came with the open. A real reader might click one link, maybe two, with gaps between them. A scanner hits everything in one pass.
If your tracker shows an open plus five link clicks all in the same second, that's not a hot lead speed-reading your email. That's malware-checking software doing its job. Treat it as noise.
Step 3: What email client is on the other end?
Ask what the recipient reads email with. If it's Apple Mail, be skeptical of every open.
[Apple Mail Privacy Protection](/glossary/apple-mail-privacy-protection) pre-fetches tracking pixels on Apple's servers, so every Apple Mail recipient looks like an opener even if they never read a word. And Apple Mail is roughly 58 percent of email client share, per Litmus. This isn't a corner case you can shrug off. On most lists it's the majority.
Gmail is its own puzzle: images route through Google's proxy, and a proxy open needs the full signal picture before you count it or toss it. Neither blind counting nor blind discarding gets it right.
Step 4: Did they come back?
Watch for repeat opens. This is the check that works in your favor.
Machines touch your email once, when it arrives. People come back. Someone who opens your proposal Tuesday morning and again Wednesday night is real, engaged, and probably worth a call. Real people open in human patterns: normal hours, real devices, returning to re-read the stuff that matters to them.
A second open at a human hour is close to proof of life.
What does a clearly human open look like?
Worth pausing on, because it's the pattern you're hunting for. Your email goes out at 9:12 a.m. Nothing for two hours. Then an open at 11:34 a.m., from a normal email client. One link click a minute later, on your pricing page. Another open that evening at 8:15 p.m.
Every part of that sequence is human. The delay, the single deliberate click, the return visit after dinner. No scanner behaves that way, and no pre-fetch system comes back for seconds. When you see this shape, follow up fast. This is the signal all the filtering exists to protect.
Step 5: Let a tracker do this for every email
Here's the honest limit of steps 1 through 4: they don't scale. You can hand-check the opens on one big proposal. You can't hand-check 80 sends a day.
That's the job of confidence scoring. Outsolvi runs these checks on every open automatically, using request timing, network origin, and client signature, and grades each one from Tier 1 (high-confidence human) to Tier 5 (bot or scanner). Opens under a 25 percent confidence floor never reach your counts. The rules are documented on the [methodology page](/methodology), and the [confidence scoring](/features/confidence-scoring) page shows how tiers appear in the app.
You can audit any decision. The [Diagnose view](/features/diagnose-open-evidence) lets you click an open and see exactly why it counted or got filtered. The model agrees with human-rated classifications 95-98 percent of the time, with Tier-1 false positives under 2 percent.
What now?
Run the five checks on your last important email. If you find instant opens and all-links-at-once bursts, you've been following up with software.
Then decide if you want the checks running full-time. The [comparison hub](/compare) shows how trackers differ here. Outsolvi is $7 a month billed yearly, 14-day free trial, no credit card. Send for two weeks and see which of your opens survive.