Diversification Without Chaos: How Agencies Manage Multiple Channels at Scale

We know you’re using other traffic sources too – yes, we definitely know 🙂

And that’s completely normal: we’ve never met anyone who believed in sticking to just a single source. So the problem is not how many sources you juggle – but how efficiently you can control them.

If you’re a single buyer working with a couple of offers, that might not be a big problem. But when you are an agency with multiple clients, and each client has several products, with various traffic sources used for each, this can become a complete mess.

We asked Omer Vardi, Media Buyer at Z2A Digital, and Alexander, independent media buyer working across multiple traffic sources, to share their real-world approach and best practices.

---

Traffic Diversification Basics: What Does It Look Like?

There’s hardly a media buyer who doesn’t diversify traffic. Even if your favorite traffic source delivers great results, scaling still requires finding new audiences across different formats and platforms. 

As Alexander explained it: 

Omer agrees: ‘We always mix traffic and try different sources – we test all the time and never rely on just a single platform. There are networks where we can pay around $1 per install, but sometimes it’s better to go with another one – even at $7 per install – because the users are higher quality. In the end, it’s about balancing cost and quality across different sources.’

So, the main diversification triggers are pretty simple: 

Enough with whys, let’s jump to hows: here’s what a workflow looks like – on the example of our respondents’ best practices.

---

1. Begin with a basic set of sources 

You don’t need to go all-in and launch campaigns on every available platform. In most cases, agencies and single buyers begin with several core, well-tested sources. For example, here is how this starting setup looks at Z2A:

Smaller agencies, teams, and buyers follow the same logic – for example, Alexander simply launches a campaign, and then adds other sources when he wants to scale or lacks volume.

---

2. Getting the first data

As Omer puts it, the point of the first results is not aggressive optimization. Instead, buyers gather the data to see how the sources perform, check the first volumes, and make sure the quality is good enough, and decide if there’s a window for scaling.

---

3. Scaling what works, testing what’s new

According to Omer and Alexander, the flow is pretty standard at this step:

• You scale the best performing sources
• You begin testing the new sources with the low volumes

Omer:

So, you don’t need to chaotically add more budgets to all the sources, but scale gradually.

---

4. Testing similar sources for failed bundles 

Another important pattern Omer highlighted: if the performance is not that great at the main trusted source, it mainly won’t work well at the other similar platforms, too.

It doesn’t mean you don’t need to test at all. However, in this case, testing on another network is not about fixing what doesn’t work at the main source – but about making sure you have squeezed everything out of all the available platforms.

As you see, the diversification flows are pretty standard and straightforward. The issue begins on a completely different side: when you manage all the sources that you add and remove.

---

Data, Attribution, and Chaos: The Main Problems of Diversification 

When you get more and more traffic sources, the problem of ‘where to buy traffic’ doesn’t exist. Instead, you face a different kind of challenges:

• How to compare performance
• How not to lose control of the campaigns
• How not to confuse all the campaigns and settings
• How not to break settings and postback
• How to understand how each source performs, after all?

Even for experienced teams, adding more sources quickly turns into a problem of control:

Omer:

And this isn’t just an agency-level issue.

A solo media buyer running a much simpler setup faces a different — but equally limiting — problem:

So, traffic diversification doesn’t break at the level of traffic, but at the level of control.

To understand why, let’s look at more details from the buyers’ experience: what actually goes wrong when the number of traffic sources grows?

---

1. Loss of control over campaigns

As more platforms are added, keeping everything under control becomes increasingly difficult. Campaigns start running simultaneously across multiple sources, and it becomes harder to monitor each of them. Some get less attention than others, some are checked too late, and some are simply forgotten.

The result? Performance drops, and not due to the traffic quality, but to missed actions.

---

2. No clear attribution across sources

Another issue appears when multiple sources run at the same time — especially on the same GEO. Traffic begins to mix, and while you still see conversions, their origin becomes unclear.

You may clearly understand how much you spend in each network, but seeing which source actually drives results becomes much harder. So, no clear attribution makes the campaign analysis more of a guesswork.

---

3. Broken tracking and data gaps

Every additional traffic source increases technical complexity: each requires proper postback setup, tracking parameters, and data integration.

In reality, things don’t always work perfectly. Data may not pass correctly, metrics may not match across platforms, and updates may be delayed. Teams often end up spending a budget just to understand how data flows.

---

4. Over-diversification without a system

As diversification grows, another problem emerges — trying to do too much at once. Adding multiple sources without a clear process quickly leads to fragmented attention and inconsistent optimization.

Instead of improving performance, this creates confusion. Campaigns are launched, but not properly followed up on. Decisions are delayed, and overall efficiency drops. At this point, diversification stops being a growth strategy and becomes a source of operational risk.

---

5. Manual work doesn’t scale

Without automation, every new source adds more manual work. Campaigns need to be checked, placements reviewed, performance monitored, and settings adjusted — often on a daily basis.

At first, this may seem manageable. But as the number of sources grows, the workload increases significantly. What used to take minutes turns into hours, and the process becomes repetitive and difficult to sustain.

---

How To Get Full Control Of the Traffic Diversification Process?

We already mentioned the workflow for diversification, and getting it right is crucial for maintaining control.

A mindset like ‘I’ll scale what’s already working and test new platforms gradually’ – instead of ‘I’ll just launch small tests here and there’ – helps avoid two common mistakes:

What also matters – and can be called a core principle – is automation and transparency. Here’s what experienced buyers recommend:

And now, to the details.

---

1. Start with a single source of truth for all campaigns

The first step is not automation – it’s visibility.

Before scaling diversification, teams need one place where they can see:

• All campaigns
• All traffic sources
• All key metrics

In practice, this often starts very simply:

Omer:

Here is how this Google Sheet might look in structure:

Campaign ID

Campaign Name

Traffic Source / Network

App / Bundle ID

Campaign Type / Category

Manager / Owner

Status (Run/Pause)

Tracker Campaign ID / External ID

CPI

Spend

Revenue

Profit/Loss

Notes / Changes

Note: In-house solutions have an obvious advantage over the ready tools, but there are still plenty of trackers out there that solve the attribution problem – for example, Scaleo, Voluum, or Binom.

---

2. Connect data automatically

Manual tracking quickly becomes unreliable as more sources are added. That’s why the next step is automation of data collection:

Omer:

This solves several problems at once:

• Removes the need to manually check multiple dashboards
• Reduces human error
• Makes source comparison easier

---

3. Create a clear structure for tracking

One of the biggest blockers in diversification is unclear attribution.

As Alexander admits, it’s impossible to differentiate traffic from two or more networks if the campaign is running in the same geo. Due to the traffic mix, there is no way to evaluate how exactly the second network works.

To avoid this, you need to structure tracking from the beginning – here’s, for example, how Omer’s team does it:

---

4. Don’t automate everything at once

Interestingly, even advanced teams don’t automate everything immediately. As Omer shared, the automation comes only after you make sure the source works fine. So, the process involves four steps:

1. Test manually
2. Confirm performance
3. Integrate into the system
4. Automate

This helps to avoid wasting time on sources that won’t scale anyway.

---

To Sum It All Up: Systems First, Then Scale

One of the clearest takeaways from both interviews is this: diversification must follow systems. It’s not hard to add multiple traffic formats and sources: the biggest question is how to handle them.

When there’s no structure in place, things tend to break quickly: new sources are added faster than you can properly manage them. The opposite happens when you build a standardized workflow and automate what can be automated.