Your DMs didn't drop. Meta's measurement did.

If you've been looking at your Meta ad reports in 2026 and feeling like the numbers tanked, you're not imagining it. The reported numbers did drop for a lot of accounts in January and February. The actual performance, in most cases, didn't.

This is one of those moments where what Meta is showing you and what's actually happening in your DMs are two different stories. Here's the plain English version of what changed, what it means for your reports, and what to actually pay attention to.

What changed in January 2026

On 12 January 2026, Meta deprecated two attribution windows that had been available in reporting for years: 7-day view and 28-day view.

Attribution windows are how Meta decides which ad to give credit to when a conversion happens. The "view" windows specifically credit ads that someone saw but didn't click on. So if someone scrolled past your ad, didn't click, but DMed you three days later after thinking about it, the 7-day view window would have counted that as a conversion attributable to your ad.

Those view windows are now gone. Meta still tracks click-based attribution, but it no longer credits ads for view-through conversions in standard reporting.

Why this matters for tattoo artists specifically

View-through attribution mattered more for tattoo work than it did for most industries. Tattoo buyers don't usually click your ad and DM you in the same minute. They see the ad, screenshot it, sit on it for a day or three, then open Instagram directly, navigate to your profile, and DM you.

Under the old system, Meta could see that pattern and give your ad credit. Under the new system, it can't. Those conversions still happen. They just don't show up in your ad report any more.

If you ran roughly the same ad spend in December 2025 and February 2026 and got similar DM volume in your inbox, but your Ads Manager report shows fewer "messaging conversations started" in February, that's the attribution change, not a performance drop.

The reframe

The way to think about this: your reported numbers are now more conservative. They show fewer conversions than the old system would have. But the actual conversions, the actual DMs in your inbox, haven't changed.

This is a measurement change, not a performance change. The plumbing inside Meta is the same. Your ads still reach people, still get seen, still drive DMs. Meta just isn't claiming credit for as many of them on paper.

Some industries had hard reactions to this. Performance teams panicked, paused campaigns, reduced budgets. That was a mistake in most cases. The right reaction was to recalibrate expectations on the reported numbers, not to change the underlying campaign.

What we adjusted internally

When this rolled out, we made three changes to how we report and operate:

1. We don't compare reported "messaging conversations" pre and post January 12 as if they're the same metric. They aren't. The old metric included view-through. The new one doesn't. Comparing them directly makes January 2026 look like a cliff. It wasn't.
2. We weight inbox volume more heavily. Inbox DMs are the source of truth. They always were, but the reported numbers used to match closely enough that we could lean on them. Now we ask clients to give us a rough count of inbox DMs and we use that as the primary read on performance.
3. We give more time on new campaigns. Because reported numbers undercount, a campaign can look weaker on paper than it actually is. We extend test windows and ask for more inbox data before deciding to kill or change a campaign.

What you should look at instead

If you're running ads yourself or auditing what's happening, here's what we'd suggest looking at, in priority order:

• DMs in your inbox. This is the only metric that matters in the end. Count them weekly. Track whether they're going up, flat, or down compared to the previous quarter.
• Profile visits. Visible in Instagram Insights. A good proxy for whether your reach is healthy.
• Reach and frequency in Ads Manager. These didn't change. If your ads are reaching the same number of people at a similar frequency, your top of funnel is still working.
• Click-through rate. Still trackable, still meaningful. If CTR is healthy, the creative is working even if the conversion column looks lower.

The "messaging conversations" column in Ads Manager is still useful, but only as a directional signal. Don't take it as the full picture.

The honest caveat

Some accounts did experience real performance drops in early 2026, separate from the attribution change. Q1 saw some genuine market softness in a few regions. New competitors entered some markets. The Andromeda algorithm continued to favour creative variety, so accounts running the same three creatives for months felt fatigue more sharply.

So the answer isn't always "it's just attribution." Sometimes there is a real issue underneath. The point is you can't tell which one it is just by looking at the Ads Manager dashboard. You need the inbox count, the profile visit trend, the creative refresh history, and the spend trajectory to know whether something's actually wrong or whether the reporting just got more conservative.

When we audit an account that's reporting lower conversions in 2026 versus 2025, we look at both. Attribution change explains some of the gap. Real performance shifts explain the rest. Treating it as one or the other is how decisions get made wrong.

What this means for reports going forward

If you're getting reports from someone managing your ads (us or anyone else), expect 2026 numbers to look softer than 2025 numbers on the messaging conversation count, even when the underlying performance is the same or better. Anyone who's not flagging this and explaining it is either not paying attention or hoping you won't notice.

The honest framing: ad performance is now measured a bit more like real-world cause and effect. View-through credit was always a stretch, and Meta dropping it is closer to the truth, just less flattering on the report.

The bottom line

Don't panic at 2026 numbers. Don't compare them apples-to-apples with 2025. Look at your inbox, not just your dashboard. And if your ad manager isn't talking to you about attribution changes when they affect your reporting, that's worth a conversation.

If you'd like us to look at your account and tell you whether the drop is measurement or performance, that's part of what we do on the trial. Apply for a fortnight on us, no management fee for the trial.