# Traffic Distribution AI: Overview

Optimization of campaign funnels is always a challenge in advertising marketing. Monitoring campaigns and adapting them to changing environments is time-consuming and quite often full of drudgery when doing it on a daily basis. This is where Voluum AI comes into play.

The Traffic Distribution AI feature automatically adjusts weights of paths, landers, and offers in the campaign funnel every minute. The weight value is used to determine traffic distribution going towards specific elements. By periodically adjusting weights of campaign elements Traffic Distribution AI changes the flow of traffic to the most optimal one. In order to take advantage of fresh data, they are based on the last 24h data set.

### Danger

Pricing plan availability

Note that this feature is available only on Profit pricing plan and above. A trial option is available for some features on a lower pricing plan. To request a trial, click the icon in the top right corner of the screen, select the feature that you want to test and click the Request trial button.

#### How Does It Work?

At the very beginning when you start creating the campaign, you will see a linear distribution of the weights, but then the values will evolve with each calculation run. As a result, the weights will vary depending on the performance of each campaign element. Note that adding a new element results in traffic being retargeted to the newly added element what allows Voluum to estimate its performance. However, taking into account that the calculation of the weights happens every minute in Voluum, all weights will be adjusted accordingly while the data is gathered.

All the calculations are based on the number of conversions that are registered in Voluum, so you need to track conversions to benefit from this feature.

#### Traffic Distribution AI for Different Elements

The scope of the Traffic Distribution feature includes traffic optimization for the following campaign elements:

• Offer optimization: Optimize weights of offers in a path.

• Lander optimization: Optimize weights of landers in a path.

• Path optimization: Optimize weights of paths in a set of paths, such as paths inside rule-base paths set or default paths set.

#### Using the Traffic Distribution AI feature

You can turn on AI optimization for all or for some elements only, with calculation method selected independently for paths and for lander and offer. To learn how to enable this feature, read the Using Traffic Distribution AI article.

#### Calculation Methods

In order to give your campaigns the best chance at success, you can always set a metric you prefer the most. The automatic weights might be calculated based on the Return of Investment (ROI), Earnings per visit (EPV), or conversion rate (CV) data. The default metric for the calculations is ROI and it is set when you select the Auto option in the platform. You can change it at any time for paths, and separately for offers and landers. Note that the same metric will be applied for both landers and offers, and you cannot select different calculation methods for landers and offers.

Automatic selection of a calculation method

The AI model in Voluum is adjusted to the fast-changing environment, so whenever it happens that there is no data even for a given element, the calculations continue by automatically switching to the next metric in the model's sequence:

• If there are cost and revenue reported for every element in the path, the calculations of the weights are based on ROI.

• If the revenue is reported for every element in the path but there is no cost information in Voluum, the calculations are based on EPV.

• If there is no revenue and cost, the calculations are based on CV

This approach ensures that even if you select one of the available metrics but somehow the metric value is not tracked in Voluum, Traffic Distribution AI will still be applied and the optimized weights will be calculated.

Overriding manual selection of a calculation method

Take into consideration that the goal of Traffic Distribution AI is to maximize you profit, so manually selected metric may be overridden if data calculated using different metrics suggest higher profitability of a given element. For example, let's assume that you have selected the EPV metric as a calculation method for your offers. One of your offers has a low EPV value, but the CV value is higher than for any other offer. In that case, the Traffic Distribution AI will push most of the traffic to the offer with the highest CV value, but also some traffic to offers with high CV.