A sudden drop does not always mean the page broke. Work through tracking, speed, traffic and the offer in that order before changing anything.

Main points

  • Check conversion tracking statuses and test a lead yourself before touching the page; misconfigured tags make healthy funnels look broken.
  • Safari and Firefox tracking prevention plus ad blockers can remove 20 to 40 percent of browser-side pixel events before they reach the platform.
  • 70 percent of consumers say slow loading times reduce their willingness to buy, so speed checks belong early in any diagnosis.
  • In one analysis, 54 percent of ecommerce sites with conversion drops also saw rising bounce rates from poor traffic sources.
  • Site redesigns and CMS migrations are the most common cause of broken conversion tracking; a change log makes the next drop easy to trace.

A landing page that converts steadily one week and barely at all the next usually has one of four problems: tracking broke, the page got slower or harder to use, the traffic changed, or the offer changed. The order of checks matters. Testing tracking first keeps you from rebuilding a page that was never broken, and blaming the offer before checking your own edits can send you chasing an advertiser’s problem that you caused yourself. Work down the list in sequence and resist the urge to pause every campaign on day one.

First, Confirm the Drop Is Real and Not a Tracking Problem

Start by ruling out measurement error and calendar effects. Compare the weak period against a similar timeframe, such as week over week or the same weekdays the month before, rather than a slow holiday stretch against a promo-driven peak. Day of week and seasonality both shift conversion rates without anything on the page changing.

If you advertise on Google Ads, open the Conversions page and check the Status column on each conversion action. Active means recording normally. Misconfigured means conversions have stopped recording entirely, usually because of a setup error or a broken tag. Needs attention flags problems that have not fully stopped recording. Awaiting conversions means nothing has recorded in the last 7 days, which is typical for actions created under 48 hours ago or campaigns that are paused. After you fix a tag, allow up to 48 hours for the status to update before you draw conclusions.

Check Your Postbacks, Pixels and Sub ID Data

Once the platform status looks clean, test the pipeline yourself. For server-to-server postbacks, submit a test lead and confirm the conversion fires on the network side and lands in your tracker with the right payout. If the handoff itself is unfamiliar, our guide to postback URLs walks through the setup.

Pixel-based tracking fails differently. Browser-side events are vulnerable to tracking prevention and ad blockers: one analysis of Meta pixels found that Safari and Firefox tracking prevention, along with ad blockers, remove 20 to 40 percent of browser-side events before they reach the platform. If a pixel looks quiet, check whether the event fires in the platform’s test tools, and whether a server-side complement exists with a shared event ID so nothing gets counted twice.

Google Ads adds its own requirements. Click trackers and redirects must pass the GCLID through to your landing page, cross-domain linking must carry the click identifier if the conversion page sits on a different domain, and tags fired inside an iFrame will undercount.

Sub IDs are the fastest diagnostic here. They break reported numbers down by placement, ad or geo, so you can see whether the whole funnel went quiet or one segment dragged the average down. Our article on sub IDs and click tracking explains how to structure them.

Rule Out Speed and Mobile Display Issues

Most lead traffic is mobile, and small performance changes show up in conversion rates before they show up anywhere else. One survey found that 70 percent of consumers say slow loading times make them less willing to buy. Run the page through a speed testing tool and compare the result against an older report if you have one, since a script added during a redesign can add seconds to load time.

Then test like a visitor. Open the page on a real phone over cellular data, click every button and submit the form. Common breakage includes a field that no longer validates, a button hidden under another element on small screens, or a consent checkbox that fails silently. Session recordings and heatmaps can show where visitors stall or exit, which beats guessing.

Compare Traffic Mix Before and After the Drop

If tracking and the page check out, the traffic itself may have moved. In one analysis, 54 percent of ecommerce sites that saw conversion drops also saw bounce rates rise from poor traffic sources. Pull before-and-after numbers segmented by source, device, geo and placement, and look for:

  • A new placement or audience expansion bringing in lower-intent clicks
  • A shift in the device mix, especially a jump in mobile share
  • One geo or app placement converting far below the rest
  • Click patterns that look automated, such as uniform session lengths or bursts of clicks with no form interaction

A flood of low-quality traffic depresses the blended rate even when every other part of the funnel works. If the clicks look automated rather than merely uninterested, invalid traffic is its own investigation and deserves a separate pass.

Audit the Offer: Payout Changes, Caps and Scrubs

When your own funnel looks intact, look upstream. Networks and advertisers change terms mid-run: payouts get cut, daily caps fill early in the day, and leads get scrubbed for quality after the fact. Any of these can lower your effective conversion rate even when form submissions have not moved. Ask your affiliate manager directly whether anything changed on their end.

Attribution settings matter too. If the advertiser or network shortens the conversion window, late-converting actions quietly drop out of reports even though the actions still happened. Our explainer on conversion windows and attribution covers how window length changes the numbers you see. Also check the post-click flow: a longer advertiser form, added verification steps or new drop-off points between their pages will show up in your numbers. And if a promo or discount period just ended, some decline is expected rather than diagnostic.

Review Recent Page Edits and Form Changes

Site changes are the most common cause of broken conversion tracking in practice. A redesign or CMS migration can leave tags behind: the new template loads without the tag manager container or the pixel, everything else on the page works, and nobody notices until the data goes flat days later. Container republishes cause their own failures, such as a conversion tag being overwritten or a trigger whose URL pattern no longer matches the actual conversion page.

Form edits deserve special scrutiny. Every added field costs submissions, and a required field visitors do not understand can stop a form cold. If a specific edit lines up with the start of the drop and you can revert it cleanly, roll back first and diagnose calmly afterward. Keep a change log of page edits, tag publishes and offer updates so the next drop takes minutes to trace instead of days.

Pro tip

Watch form submissions and tracked conversions side by side. When submissions hold steady but conversions vanish, the page is fine and the tracking is not.

Rebuilding the Baseline: When to Test and When to Roll Back

Once you have found the cause, rebuild in order. If the fix was technical, confirm conversions record for a few days before resuming full spend. If the cause was traffic, cut the weak segment or tighten targeting and watch the blended rate recover instead of redesigning a page that was fine. If the cause was the offer, decide whether the new terms still clear your margin, or whether the same page should carry a different offer.

After that, test one change at a time and only once the baseline is stable. Changing the headline, the form and the traffic source in the same week guarantees you will never know which move mattered. Expect recovery to take days of data, not hours, especially on lower-volume pages.

Diagnosis is mostly about order, not effort. Work from tracking outward to the offer and you will usually find the one thing that moved. If the trail ends at the page itself, our comparison of long and short landing pages can help you decide what to change first.

Frequently asked

Should I pause my campaigns as soon as the conversion rate drops?
Not right away. A tracking failure can look identical to a page problem, so verify tracking and compare similar timeframes first. Pausing everything also destroys the data you need to diagnose the issue.

How do I tell a tracking problem from a real conversion drop?
Compare tracked signals against untracked ones. If form submissions or server-side leads hold steady while platform-reported conversions fall, the page works and the measurement chain is the suspect.

How long should I wait after fixing a tag before judging results?
Give it a few days of data. Google Ads conversion statuses can take up to 48 hours to update after a tag change, and low-volume pages need longer before the numbers mean anything.

Can traffic quality alone cause a conversion rate drop?
Yes. The blended rate falls when low-intent traffic enters the mix even if the page and funnel work exactly as before. Segment by source, device and placement to find the weak slice.

This article is general information for marketers, not legal advice. Advertising, privacy and consumer protection rules vary by country and change over time, so check the current rules and your network’s terms, and get legal advice where needed before running campaigns.