Three Years of Wrong Audio Buys: From a JBL Go 3 to an NAD M23, Here's What I Learned

The Wedding

September 17, 2023. A backyard wedding with 78 guests. I was the designated "tech guy" responsible for the music.

On my table: a JBL Go 3 portable bluetooth speaker, a generic USB loudspeaker I'd borrowed from my office, and an audio console mixer I'd bought the week before and barely understood. The sound died within the first three meters. The Go 3 is a brilliant little speaker for a bedroom, a hotel room, or a quiet park bench. It isn't built for outdoor ceremony coverage with 78 conversations stacked against it. Routing everything through the mixer—because that's what professionals do, right?—just added noise and confusion to an already losing setup.

The bride's father walked over before the ceremony started. He didn't say a word. He looked at the table, then at me, then back at the table. That look said everything: "You had one job."

I deserved it. I'd been making the same class of mistake for three years. The wedding just made it impossible to ignore.

How I Became "The Audio Guy"

Let me back up. I'm not an audio engineer. I'm not a musician. I was the person who could connect a laptop to a conference-room TV without calling IT, and that was the entire qualification.

In 2020, a friend asked me to handle music at her birthday party. "You know about this stuff, right?"

Sure. I didn't, but sure.

I brought my desktop USB loudspeaker—one of those cheap models with a single 3.5mm input and a volume knob that crackled when you moved it. On my desk, it was fine for background YouTube. At a party with 20 people, it had no chance. I turned the knob up, and the sound didn't get louder. It just got more distorted. The guests leaned in, squinting at me, pretending to be able to hear the songs.

That night, a guy told me something I still remember:

"RMS is what actually matters. Peak power is marketing."

I didn't look up what RMS meant that night. I was too busy being embarrassed. But the line stayed with me.

The JBL Go 3: A Great Speaker in the Wrong Place

By 2022, I owned a JBL Go 3. I bought it because of the IPX7 rating (IEC 60529: immersion in one meter of water for 30 minutes) and because it was small enough to throw in a rucksack. And it held up—I've spilled coffee on it, dropped it on gravel, taken it out in light rain. It still works today. Dollar for dollar, it's one of the best purchases I've ever made.

But here's the trap. A great speaker in a small category can convince you it's a great speaker in every category. I started using the Go 3 for everything: kitchen, garage, car, small meetups. Slowly, I began to believe it was "loud enough for most things."

It isn't. It can't be. The Go 3 puts out roughly 4.2 watts RMS. Four point two. That's not a criticism of the product—it's the spec sheet. A 4.2W RMS speaker is designed for close-range listening, not for projecting sound into a group. At a house party in May 2023, I learned this the hard way a second time: 25 people, one small speaker on a table, everyone leaning in. The bass simply didn't exist, and the top end sounded like a tin can.

The problem was never the Go 3. The problem was me, using a personal speaker as a room speaker.

The "Go Pro" Mistake

After that house party, I decided I was done with consumer gear. I was gonna go professional. That is exactly when I set up my next three mistakes.

First, I bought an audio console mixer. Not because I had a specific routing problem—because I assumed that professionals use mixers, so mixers make things sound professional. Second, I bought an NAD M23 Masters Stereo Power Amplifier. I'd read the reviews: 200W into 8 ohms, built around the Purifi Eigentakt Class-D module, with measured distortion figures that embarrassed amplifiers five times its price. I thought raw power would fix everything.

It didn't.

The NAD M23 and the speaker mismatch

The NAD M23 is an honest product. I have no complaints about the amplifier itself. But I connected it to a pair of old bookshelf speakers rated at 86 dB sensitivity (1W at 1m), with an impedance curve that dropped below 4 ohms in the bass. In a living room that was too large for them.

Amplifier power doesn't fix speaker limitations. Power gives you headroom, but the sensitivity of the speaker determines how efficiently watts become sound pressure. Feed 200 watts into inefficient speakers and you get volume, sure, plus heat, and a growing disappointment. A high-sensitivity speaker (say, 92 dB at 1W/1m) would have reached the same loudness with a fraction of the power.

I did not understand this in 2023. I believed "more watts = more better."

The mixer level problem

The audio console mixer taught me a different lesson: signal levels. Professional mixers typically run at +4 dBu line level. Consumer gear often runs at -10 dBV. When you connect consumer audio into a pro mixer, you get either a weak signal, or you push the gain up and you get noise, or you push even further and you get clipping. I managed to hit all three, in various combinations, during the same week.

None of it was the mixer's fault. The fault was mine: I bought a tool without learning how it wanted to be fed.

And by September, when the wedding came, I still didn't trust the mixer enough to make it the centerpiece of a live setup—so I showed up with the Go 3, the USB loudspeaker, and the mixer that I used as a glorified volume knob. The combination failed gracefully, like it was born to fail.

The Handlebar Epiphany

A year later, I bought a JBL Wind 3S Slim Bluetooth handlebar speaker for my motorcycle. Not for my living room. For the bike. It's a compact, weather-resistant speaker with a mounting clamp, designed to sit on handlebars and be heard over wind and engine noise. It has a specific job, and its design matches the job.

And you know what? It works. It simply works.

Seeing the Wind 3S succeed right next to my expensive, disappointing home setup finally made the pattern obvious. I wasn't bad at choosing audio gear. I was bad at defining the job. The Wind 3S was selected for a concrete scenario. The NAD M23 was selected because I wanted "a serious upgrade"—which is a feeling, not a specification.

When you pick gear for a specific job, it works. When you pick gear to feel like a professional, it never does.

What I'd Tell Anyone Starting Out

Here's the checklist I wish someone had given me in 2020:

1. Start with the job, not the product. Write down how many people, what space, what distance. A 4.2W RMS speaker serves one person at close range. A real event speaker starts around 50W RMS and goes up. "Loud" isn't a measurement. Listener count and square meters are.
2. Check the numbers that actually matter. RMS power, not peak. Sensitivity in dB SPL (1W/1m). Impedance matching. The US FTC made RMS power a mandatory rating standard for home amplifiers back in the 1970s for a reason: peak numbers are hype, and RMS is what you can use.
3. An amplifier doesn't fix weak speakers. It's one link in a chain. Match the amp, the speakers, the room, and the content type. The NAD M23 is a superb amplifier—once I connected it to speakers that could do it justice.
4. Know your signal chain. Consumer level into pro level without a converter is a recipe for noise and clipping. A mixer is a routing and control tool, not a magic sound-enhancer.
5. Read the wireless spec. Bluetooth has evolved: 5.3 offers better range and stability than 4.0 ever did, and low-latency codecs like aptX Low Latency exist for monitoring and video. But "it has Bluetooth" isn't a spec. Check the version and the codec.

The Industry Has Moved

One reason I kept failing was that I was operating on an outdated mental model. I thought of small Bluetooth speakers as toys and giant Class-D amps as risky, unproven technology. Both ideas are stale.

What was best practice in 2020 may not apply in 2025. In 2020, a Class-D amplifier with the measured performance of the NAD M23 was rare and expensive. Now Class-D is the default, and it's exceptional. Bluetooth was once a convenience option for background music; with LDAC and aptX Adaptive, it's now a legitimate choice for critical listening. Even the cheap USB loudspeaker has improved—though it's still a desk tool, because the physical constraints of USB power and tiny drivers haven't changed all that much.

The fundamentals haven't changed: speakers move air, and you judge the result with your ears. But the execution has transformed. Getting good sound is no longer about spending more money. It's about matching the tool to the task—a lesson that cost me about $3,400 to learn.

That Wedding Led to This

I kept a list of every bad audio purchase I made between 2020 and 2024. There are nine entries. Together, they cost me roughly $3,400. Some of that gear I still use and love—the Go 3 has earned its place a hundred times over. The Wind 3S is still mounted on my bike and still performs. The NAD M23 is now driving a pair of high-sensitivity floorstanders in my listening room, and it sounds absolutely wonderful.

The difference wasn't the gear. It was my approach.

If you're about to buy any audio equipment for a room, a bike, a desk, a van, or an event, write the job description first. Then choose the tool that fits the job. The bride's father and I actually laugh about that wedding now—I ran into him at a grocery store last year, and I apologized properly.

But the better version of that apology is the checklist above. Someone else should learn from my mistakes instead of paying for their own.

Learn what RMS means first. The gear comes after.