Ads not converting

Your Offer and Landing Page Are Why Ads Don't Convert

Bradley J Simons
Bradley J Simons
Founder of VLVTN · runs paid Meta ads for his own releases as Babbage
Updated 2026-07-26
5 min read
The short answer
Sometimes the ad works, the person taps, and you still lose them. That gap can be the offer or the landing page. Speed, button count, platform order, artwork weight, and the match between the ad and destination are testable hypotheses. Change one variable, measure DSP click-outs per page view, and keep the result the data supports.
Key takeaways
  • Use valid page visits and tracked DSP click-outs to test a landing-page hypothesis. Raw ad clicks cannot isolate the cause.
  • Treat button count and platform order as hypotheses, not universal rules.
  • Use 2.5 seconds or less at the 75th percentile as the LCP performance target.
  • Oversized artwork can become the largest contentful element and delay LCP.
  • Change one landing-page variable at a time and measure DSP click-outs per page view.

The click landed and you lost them anyway

Most of the "my ads aren't converting" advice you read is about the ad: the hook, the audience, the creative. That's usually right, because the ad is usually the problem. But there's a quieter failure that happens after someone taps. They liked the ad enough to click, they reached your smartlink, and then they bailed before they ever got to Spotify. You paid for that click and got nothing for it.

Start with the full funnel: ad click, valid page visit, and tracked DSP click-out. Verify the event first. If visits arrive and the click-out rate is weak, the destination is a reasonable hypothesis. Hold the campaign steady, change one page variable, and compare it with the baseline before assigning the cause.

Note
This guide is about the destination, the offer and the smartlink itself. It is not about whether your pixel is firing. If conversions aren't showing up at all and you think it's tracking, that's a different fix: why your pixel isn't tracking conversions.

Measure speed before you blame it

Largest Contentful Paint (LCP) measures when the largest image, text block, or video in the viewport finishes rendering. Google's current Core Web Vitals guidance calls 2.5 seconds or less good at the 75th percentile, split by mobile and desktop. That is a performance target, not a promised conversion lift.

VLVTN records real-user load time, LCP, FCP, and TTFB. Read those beside page views and DSP clicks. If slower sessions also abandon more often, optimize the measured bottleneck and run a controlled before-and-after or A/B test. The official definition and threshold are in Google's LCP guidance.

Test button count and order separately

There is no defensible universal answer to how many DSP buttons a smart link should have. A smaller set may reduce scanning, while a larger set may serve listeners whose platform would otherwise be missing. The result depends on the audience and campaign.

Test count first: keep the artwork, copy, traffic, and order fixed; compare the current destination set with a smaller relevant set. Then test order in a separate experiment: keep the same buttons and move only the campaign's primary DSP to the first position. The primary metric is DSP click-outs divided by landing-page views.

Do not change count and order in the same test. If the variant wins, you will not know which change caused it. VLVTN's A/B variants can split traffic, and the sample-size calculator gives a traffic target before the test starts.

Landing page setupWhat it does to conversions
Button-count testSame page and order; change only how many relevant DSPs appear.
Platform-order testSame DSP set; move only the campaign's primary service.
Primary metricDSP click-outs divided by landing-page views for each variant.
Decision ruleUse a preplanned sample and accept an inconclusive result.
Grade your cost per conversion against the benchmarks

Clear CTA copy comes before clever copy

VLVTN supports five DSP button labels: Play, Listen, Listen Now, Stream, and Open. None is a universal conversion winner. The first requirement is that the fan can understand what the link will do from its label and context. W3C's link-purpose guidance makes that clarity an accessibility requirement, not just a copywriting preference.

Pick one plain label as the control and one genuine alternative as the variant. Keep the platform set, order, artwork, headline, traffic, and dates comparable. Judge the test by DSP click-outs per landing-page view. Do not publish a winning-copy claim until the planned sample is complete. Read the W3C link-purpose guidance.

Artwork size is a performance variable

Release artwork is often the largest visible element on a music landing page, so it can become the page's LCP element. Raster image file size grows with pixel count: doubling both dimensions creates four times as many pixels before compression. Uploading a huge source file for a small on-screen square wastes bytes.

Serve the dimensions the layout needs, provide responsive variants, compress the asset, and use a modern format where the pipeline supports it. Then confirm the result in real-user LCP rather than declaring a file-size limit that works for every network and device. Google's image-format guidance explains the pixel-count tradeoff.

Automatic routing is not automatically better

A phone type or country does not reliably reveal which streaming service a fan pays for. An automatic redirect can remove one tap, but a wrong guess removes the fan's choice. VLVTN currently shows the available DSP buttons instead of inferring a subscription it cannot know.

If you test automatic routing elsewhere, keep an obvious way to choose another service and compare successful DSP click-outs, not just landing-page exits. A faster wrong redirect is not a better conversion.

In-app browser friction on Instagram and TikTok

Here's one that catches a lot of people and is invisible unless you know to look for it. When your ad runs inside Instagram or TikTok and someone taps the link, it doesn't open in their real browser. It opens in an in-app browser, a stripped-down web view inside the app itself. That in-app browser can fumble the hand-off when the listener tries to jump from your link into the actual Spotify app, and the listener gives up.

You can lose conversions here without ever seeing why, because on your own phone, in a normal browser, the link works perfectly. vlvtn detects in-app browsers and forces a same-tab redirect so the hand-off to the DSP holds. It's a real source of lost clicks and one of the reasons the smartlink exists.

Tip
Test your own link by tapping it from inside Instagram, not just from your phone's browser. If it stalls or refuses to open the Spotify app from there, that's the friction your listeners are hitting.

When the offer doesn't match the ad

The last one is about the promise, not the page mechanics. Your ad sets an expectation: a vibe, a specific song moment, a feeling. If the listener taps and the landing page or the track delivers something different, they bail. The click happened, the conversion didn't, and it's not because the page was slow. It's because the destination wasn't what they were sold.

The fix is alignment. If the hook of the ad is a particular moment in the song, that should be front and center, with the artwork and the framing matching the energy of the creative people just watched. The ad and the landing should feel like the same thing. The further apart they drift, the more people fall through the gap between them.

Watch out
A mismatched offer can look exactly like a tracking bug: clicks go up, conversions don't. Before you blame the pixel, ask whether the page actually delivers what the ad promised.

What the landing page can and can't fix

A better page can remove measurable friction between a valid visit and a DSP click-out. The amount has to be tested. A conversion is a tracked click-out from your smartlink to the streaming service. It is not a guaranteed stream, save, or follow.

If the tested page raises click-outs while Spotify stays flat, move to the downstream-quality question. Align the reporting windows, account for other traffic, and inspect Spotify for Artists. The page can prove the outbound click. It cannot prove listening or isolate why the DSP number changed.

If you're not sure which problem you have, the pillar guide on why music ads don't convert walks through all three non-tracking causes. If the ad itself isn't pulling clicks, start with the creative reasons or the targeting mistakes before you touch the landing page.

Use the landing-page conversion-rate guide for the exact visits-to-click-outs calculation, then read clicks without Spotify streams before drawing a downstream conclusion. When you need the tracked destination itself, create a VLVTN smartlink.

Frequently asked

How do I know it's the landing page and not the ad?

Compare valid landing-page visits with tracked DSP click-outs after verifying the event. If visits arrive and the click-out rate is weak, the page or message match is a reasonable hypothesis. Test one page change against the baseline. Raw ad clicks alone cannot isolate the landing page.

How many streaming services should I put on the smartlink?

There is no universal winning number. Include the services your audience actually uses, make the campaign's primary destination easy to find, and test a smaller set against your current set. Judge the result by DSP click-outs per landing-page view, not by preference or intuition.

Should a smart link automatically route the fan?

Only when the destination is known with high confidence and the fan can still choose another service. A device or country does not reveal a person's paid music subscription. VLVTN currently shows the available DSP choices rather than guessing a subscription it cannot know.

Does a fast landing page actually move conversions?

Speed removes a source of friction, but the conversion lift is not a universal fixed number. Google defines a good Largest Contentful Paint as 2.5 seconds or less at the 75th percentile. Measure real-user LCP and click-out rate together, then test a speed change against a control.

Which smart-link CTA gets the most clicks?

No CTA is a universal winner. The label should make the destination and action clear, then be tested against one alternative with the rest of the page held constant. VLVTN supports Play, Listen, Listen Now, Stream, and Open. Measure DSP click-outs per page view and accept an inconclusive result.

I fixed the landing page and conversions went up but streams didn't. Now what?

A conversion is a tracked click-out to the DSP, not a guaranteed play. Recheck the campaign and Spotify for Artists in matching windows, account for other traffic, and treat the gap as a downstream-quality question. The page cannot observe what the listener did inside Spotify.

Bradley J Simons
About Bradley J Simons
Founder of VLVTN · runs paid Meta ads for his own releases as Babbage

Bradley J Simons founded VLVTN and runs his own paid Meta and Spotify ad campaigns as the artist Babbage. He writes about paid music marketing from the buyer's seat, with his own money on the line.

Keep reading

Grade your own cost per conversionPaste in your cost per result and see where it lands against real benchmarks.Open the grader