Loudly review — 7/10 after a real test

Test scenario

Compact

Step 1

Is it clear in 10 seconds what the product does and who it’s for?

On the first screen, everything is immediately clear: "An AI-based music creation platform. Use various models to create original tracks, remix songs, improve sound quality…" There's no clear positioning for who exactly this platform is for; the team simply generally states "for Creators." Very convenient—the first screen immediately has a field to enter your prompt. After entering it, you'll be taken to the generator page directly at https://www.loudly.com/music/ai-music-generator where additional settings and platform features become available to you. Before you start generating, the project team will familiarize you with the service's advantages and try to "sell" it to you again. So even before registration, you become an active user of the platform and get a positive user experience.

Step 2

How many sign-up steps are there, and is a card required immediately?

Three provider options are available for registration: Google Auth, Facebook, and email/password registration. The registration window itself presents no issues. During registration, the platform asks not only for your name and email, but also your avatar. Attention to detail is appealing, but your avatar doesn't appear in the account dashboard, raising the question of why the project team requests your digital portrait at the registration level. After registration, you stay on the generation account page and receive 50 tokens added to your balance. Generating 1 track will cost you 12 tokens, at least that's what's written on the main "Generate" button. Thus, "on paper," this is one of the most generous offers in the segment—you should theoretically be able to generate 4 tracks. But the problem is that your prompt is no longer visible to you, and if you forget what exactly you planned to create, you'll need to find the appropriate tool on the left panel.

Step 3

Are newcomers guided, or dropped into an empty interface?

There's no formal onboarding, but the interface and behavioral logic itself nudges the user toward achieving first value: prompt on the main screen, seamless registration, and return to the generation account page with the saved prompt. Everything is very "smooth" and seamless. But precisely because there's no onboarding, the system decided to generate 35-second hip-hop tracks without vocals for us, essentially hinting that next you need to sign up for a paid subscription. Despite this, we were confident we were generating a song. The generous offer essentially turned into user disappointment and wasted 12 tokens on tracks that could essentially have been pulled from a ready-made library, dusted off, and given to the user for review.

Step 4

How long from “Get started” to the first useful result?

Generating first value was an unpleasant surprise! We expected to get a song with vocals based on our prompt using 12 free tokens, but got two melody options at 35 seconds each. That's not what we wanted. The service generated 2 tracks, as is traditional for this type of service, with completely different styles and sounds, which suggests this was taken from a ready-made library and we didn't get "honest" generation. Unfortunately, we can't verify this. In defense of the platform, we note that the received 35-second tracks can be downloaded in mp3 format for free to your device. Time: 6:30

Step 5

Does the product work on a different task beyond the demo case?

We approached creating second value more consciously and selected the necessary Text-to-Music generation section, entered our prompt again (fortunately we saved it), and specified that we wanted a "lyrical" style. The track price came to 20 tokens, also from the free balance received during registration. The balance remaining is 28 tokens. Surprisingly, 2 tracks were generated, but one was 35 seconds and the other 3 minutes 10 seconds. We asked the platform to create tracks in Ukrainian; the vocals were frankly poor, with a wild Anglo-machine accent. The song lyrics don't appear either when playing the track or otherwise. But we can still download the track to our device. The result didn't satisfy us. In defense of the platform, we note that the received tracks can be downloaded in mp3 format for free to your device.

Step 6

Where does the free plan end, and what breaks on unusual input?

The platform gives you the opportunity to explore music generation capabilities slightly more than competitors do. You get slightly more attempts; generating two versions counts as one and everything is predictable. Before each generation, you clearly see how many tokens will be deducted from your free balance, which is credited 50 tokens upon registration. Overall, you'll get 3-4 generation attempts. Also, on the free plan you can download the generated composition to your device. But at the same time, the service will remind you that not all features are available and you need to purchase a paid subscription.

Step 7

On the payment screen, is it clear what you are paying for?

Predictable and clear pricing system with token deduction. Plans are generally slightly better than the market average. You need to pay attention to the price/quality ratio. There are both monthly plans and annual commitment plans. There's no option to purchase tokens without a subscription.

Step 8

How do you unsubscribe, delete the account, and export your data?

You can delete your account directly from the user dashboard. The project team warns that deleting your account will also clear your track library. Upon re-registration, we were not asked to accept Google Auth conditions again, which might indicate "partial" deletion of the user from the database. However, after re-registration, we again had access to 50 free tokens. From our perspective, this is a vulnerability in competitive competition that the project team should fix as soon as possible.

Three focus questions

Focus 1

How clear is the product onboarding?

At first it seems like everything is going according to plan and everything is easy and simple. Disappointment comes when you get your "first value," which might not be what you initially needed. The primary prompt you enter on the main screen stays "under the hood" and instead of a song you get a 35-second melody. The project team needs to think about a more predictable user "immersion" logic and reconsider their "first value" acquisition strategy.

Focus 2

Is the pricing logical?

Overall, there are no questions about pricing. For $10 per month with monthly billing, you get 2500 tokens and a maximum of 7-minute tracks. Everything is well described on the pricing page, and when generating, the service shows how many tokens will be deducted for creating a composition. Conditionally, with 2500 tokens you can get 125 full tracks, since text-to-music generation will cost you 20 tokens per generation. You also have access to annual commitment plans with savings of up to 20%, but you'll need to pay a year in advance. There are no complaints about pricing; the plans generally correspond to the project segment and have competitive rates. When choosing plans, you'll need to consider not only the number of tokens for generation that will be provided with your paid subscription, but also the composition duration parameter. Also, on the basic plan at $8-10, you'll get a Personal license, while Pro is only available on the advanced plan. The difference between the two license types is described at the bottom of the pricing cards without anchor links to these sections, which adds some confusion when choosing plans.

Focus 3

Does the product suit the stated target audience?

Our expectations were not met. Generating a Ukrainian song for Loudly proved to be a challenging task, despite the service working quickly enough. The quality of the track, namely the vocals and words, leaves much to be desired. It's better suited for serial podcast authors, video creators, or background music. This wasn't clearly indicated on the first screen; the primary message reads: "Use various models to create original tracks…" We are not ready to recommend the project for songwriters, especially not for Ukrainian language songs.

Takeaways

01 What worked The speed of getting "first value" and the ability to generate more than 2 tracks on free tokens, unlike competitors. The ability to download the generated composition to your device on a free plan. The pricing is also attractive—easy to understand and token deduction for generation requests is predictable.

02 What annoyed The service failed to generate a Ukrainian song with good quality. There's a feeling that tracks in local languages of Eastern European countries will be low quality. Tracks are generated in two variants, but in the case of a song, the second variant was only 35 seconds, while the track itself (1st variant) was 3 minutes 10 seconds. During registration, the project "smoothly" redirected to the music generation form, but instead of a song, we got only short melodies because the text-to-music tool wasn't set by default. As a result, we got something different from what we originally wanted.

03 Who it’s not for Those expecting to get high-quality songs in local languages.

04 Fix first Predictable onboarding and song generation quality. Simplify plan comparison. Be upfront that the basic plan won't give you Pro license and tracks longer than 7 minutes.

Loudly review: what we found in testing

Summary