How We Actually Design a Church PA System

From the first conversation to the final hang — the process behind a Summit PA design, step by step.

There's no shortage of opinions about how a PA should be designed. Every manufacturer has a process. Every integrator has a philosophy. Every forum thread has seventeen contradictory recommendations.

This is ours — built from nearly thirty years of Summit projects, the hard lessons from early in our careers, and the kind of real-world experience that software models can't replicate.

It's not the only valid approach. But it's the one we've developed through enough mistakes and enough wins to believe in it.

Step 1: Understand what the church is actually after

The design process starts with a conversation, not a spreadsheet.

Before room dimensions or speaker configurations, we need to understand what a church wants their sound to feel like. What's the worship style? What's the energy level? What have they heard somewhere else that gave them the feeling they're chasing — and what have they experienced that they never want to feel again?

These aren't warm-up questions. They're the data that shapes every downstream decision.

Technical specs are surprisingly poor at capturing what churches actually want. An SPL number means different things in different rooms, with different mix engineers, and different reference points in the listener's memory. But "warm" and "direct" and "I want it to feel like a show, not like a PA in a church" — that's information we can work with.

We're also asking about the full range of use. What does a typical Sunday look like? What does the youth service look like? What happens when a touring artist comes through or a special event pushes the system harder than anything in a normal week? The design has to serve all of it, with enough headroom that the system is never operating at its limits just to handle what the church actually does.

Step 2: Match voicing to vision before touching the room

Different PA manufacturers don't just look different — they sound different. The way a line array renders the frequency spectrum, how it handles transients, where it feels warm or bright or punchy — these characteristics are inherent to the product and consistent across installations. They're called voicing, and they matter.

A church doing gospel worship and a church doing CCM and a church doing Bethel-style production are going to feel different in the room, and some PA products are better suited to some of those environments than others. This isn't a knock on any manufacturer — it's a reality of design philosophy and physics.

Before we build a model, we've already filtered the product conversation through the vision conversation. If you told us you want warmth and you want full-spectrum presence in the low mids, certain products are better starting points than others. That filter happens first.

Step 3: The room tells us what's hard

Room geometry is where the real engineering begins.

Height, length, width, seating rake, balcony configuration, balcony soffit angles, surface materials, room volume — all of it shapes what the PA needs to accomplish and where the physics are going to push back.

Consistent high-frequency coverage across a room is relatively tractable. Getting even low-mid energy — the 100 to 400 Hz range where most mixes are actually won and lost — from the front row to the back seats of a large room with a balcony is where the hard design work lives. That's the frequency range where harsh and warm diverge. Where speech intelligibility is actually built or lost. Where a mediocre design and a great design sound completely different.

Balconies deserve special mention because they change the problem significantly. A balcony creates a surface that the PA has to cover from below, usually at a different throw distance than the floor seating, with soffits that can create reflections if the angles aren't managed. A design that handles the main floor gracefully can fall apart at the balcony — or vice versa — if the geometry isn't worked through carefully.

Step 4: Model it, then filter it through experience

Once we understand the vision and the room, we build a model in design software. We look at how the system covers the space across frequencies, where the gaps are, and how different configurations address them.

The model is useful and necessary. It's also not the whole truth.

Most PA design software has historically operated on statistical modeling — a prediction of what should happen based on physical principles and manufacturer data. It can make almost any design look successful in the 2-4kHz range, which is relatively easy to predict. The low mids are harder. The real-world behavior of energy in a complex room is harder still.

We've walked into rooms with existing designs where the model looked good on paper and the room sounded wrong. The model isn't wrong — it just doesn't capture everything.

"It's really easy to make everything look red in the model and look beautiful. But there is real-world geometry and surface material that's going to impact whether this design actually translates."

After the model, we apply an experience filter. Years of installed systems, years of tuning, years of hearing what designs predict versus what they produce in practice — that knowledge shapes how we read the output of the software and where we push back on what it's telling us.

The design gets refined. Boxes get repositioned. Angles get adjusted. Questions get asked internally: does this actually work, or does it just look like it works?

The answer matters, because we're the ones tuning it after installation.

Step 5: Mechanical design before electronic correction

This is a foundational principle and worth stating directly: get the physical design right first. Electronics are for refinement, not rescue.

Every serious PA manufacturer — the ones on the big tours, the ones that everyone wants — will tell you the same thing privately. They build all the DSP and processing they can into their amplifiers and speaker management systems. But their own engineers will acknowledge: they push the mechanical design as far as they can before they lean on electronics to correct it.

There's a reason for this. You can't EQ a null. You can't use electronics to correct the comb filtering that happens when two boxes are fighting over the same frequency in the same space. You can't boost your way into coverage that the physical configuration doesn't provide.

"Get the mic in the right spot on the drum and it's going to sound amazing. You can swap in ten different microphones. But you can't EQ your way out of a bad mic position. Same principle, different scale."

Mechanical design is what determines whether the system actually covers the room. Electronics determine how well you can optimize what the physical design produces. That sequence matters.

Step 6: Design, install, and tune as one team

We design our own systems. We install our own systems. We tune our own systems. These three things belong together.

When the engineer who designed a PA is the same one standing in the room during calibration, the feedback loop is tight and honest. They know what they were trying to accomplish. They know where the compromises happened and why. They can make informed decisions about what to address electronically and what to accept as a constraint of the space.

Tuning a system designed by someone else is harder. You're interpreting someone else's intent, often without knowing the inputs they were given. You're making judgment calls without the context of the design decisions. And in cases where the design itself has problems, tuning becomes an attempt to patch something that needed to be resolved upstream.

This is also why we are wary to tune a system someone else installed. Not out of ego — but because the design dictates a much larger percentage of the experience than the tuning, and bad design can't be fixed in tuning.

What this means for brand and product decisions

After all of the above — vision, voicing, room geometry, modeling, mechanical design — the brand question becomes simpler.

We don't come in with a predetermined answer on brand. We come in with a group of products that we believe can achieve what the church is after, all meeting the baseline requirements for coverage and performance. Within that group, preference is a valid input. A production director who has worked with a specific manufacturer and knows its characteristics has real knowledge. That's worth honoring.

What it can't do is override the design requirements. If a church's preferred brand can't cover the room in the configuration they want, that's a conversation to have before anything is purchased. The responsibility to the church extends beyond the current production director to the next one, and the one after that. Context gets forgotten. The system doesn't.

"You have a preference. That preference is valid. Your church has a vision. That vision is valid. Our job is to help you find what ultimately serves both."

If you haven't already, go back and watch the acoustics episode in this series. The PA conversation only makes full sense in that context.

The best projects don't start with a purchase order. They start with a purpose. Our team has helped hundreds of churches find the sweet spot between their vision and their budget. We’d love to help.

If this sparked ideas, check out the full Gear Follows Vision episode here.