SPO and DPO: the Difference Between Two Win-Win Optimization Processes

The review topic is ways to optimize the technological processes of the advertising business, known as supply-path optimization (SPO) and demand-path optimization (DPO). Both processes benefit all the parties in the advertising chain: publishers, advertisers, and AdTech companies that provide platforms and services to users. Optimization leads to better campaign results, increased revenues, and building direct long-term relationships between sellers and buyers.

The DPO and SPO processes are part of the ad-path optimization (APO) concept. This term defines a set of actions that advertising participants may perform to improve ad inventory and increase revenues within the same effort and resources.

In this review, we will consider the features of DPO and SPO, reveal their algorithmic components, determine their differences and shared qualities, and provide links to the optimization mechanisms.

How it all began: the background of optimization for AdTech technologies

To make a long story short, the so-called “difficulties” began around 2016, when a programmatic sales solution — Header Bidding — began to gain popularity.

With header bidding, publishers' (SSPs) trading platforms could accept bids from many advertiser platforms (DSPs) at once. This significantly increased computing's operational costs and created many requests that were often duplicated.

That is how DPO and SPO emerged — as a response to the challenge of growing technology.

What is demand-path optimization (DPO), and who benefits from it?

Demand-path optimization (DPO) is a publisher-owned process that is performed through SSPs and other automated ad inventory supply tools such as platforms, exchanges, and networks.

This process primarily benefits publishers, but advertisers and third-party AdTech actors also profit from DPO.

What is supply-path optimization (SPO), and who benefits from it?

Supply-path optimization (SPO) is a mirrored process owned by advertisers and performed through DSPs and other platforms and tools for the automated distribution of advertising traffic within target audience groups.

The SPO process is most beneficial for the advertiser, but all other participants, including technical parties, also gain advantages from it.

Differences and common features of DPO and SPO

A simplified representation of how advertising players interact when SSP has built-in ad exchange

As you can see, SPO and DPO are interconnected processes, both beneficial for better and more accurate ad serving and increasing ad revenue. They were developed to improve automatic ad algorithms, which is why SPO and DPO have similar goals, but the ways and mechanics of their implementation differ.

The main parts of DPO for publishers:

• making the purchase easier for media buyers;
• increasing sales revenue and reducing supply chain costs;
• getting rid of minority or non-compliant advertisers while strengthening relationships with authorized majority partners.

The main parts of SPO for advertisers:

• finding the shortest way to purchase advertising slots to reduce transaction expenses and shorten the supply chain;
• revising the mechanism of bidding based on the campaigns’ success rates;
• identifying the most profitable partners through the markers and a comprehensive assessment of the interactions’ efficiency.

In both cases, participants must collect and analyze data for improvement purposes. The amount of data is always variable: it may be hidden due to privacy policies, missing in some cases, or not reduced to a single type, making it difficult to collect and analyze.

Types of data required for DPO:

• advertising revenue and transaction costs for selling slots;
• the volume of purchased ad inventory by each specific media buyer;
• page loading speed during interactions with traffic from different advertisers;
• computing load, defined as the number of requests per second;
• amount of links advertisers go through to buy slots, etc.

Types of data required for SPO:

• the number of SSPs, exchanges, and other participants used to place the ads;
• impressions received as a result of different postings;
• the cost of the acquired advertising space, as well as the total transaction price for the entire chain;
• results of media sellers’ testing scenarios (by the strategy);
• ranking of domains where ads are placed;
• the assessment of content’s relevance at the hosted pages and others.

In addition to automated data collection and analysis, marketing officers can be involved in SPO and DPO to make data-driven decisions. These may also be advertising managers who want to build a direct relationship with a company partner.

SPO and DPO cases in the AdTech industry

Revenue increased by 192%, effective cost per thousand impressions (eCPM) increased by 60%, and visibility increased by 55%. These are the results of demand-path optimization (DPO) conducted in cooperation between Setupad and Kyiv Post. During the case, much attention was paid, in particular, to reducing the number of SSPs that the media interacted with while selling their advertising inventory.

Reduced cost-per-action (CPA) by 28%, reduced cost per mille (CPM) by 46%, and increased number of impressions by 111%. These are the results of supply-path optimization (SPO), which was conducted due to the cooperation between PubMatic and The Goodway Group, acting on behalf of the advertiser. This remarkable result was achieved by obtaining detailed data at the log level, the key of which is the win or loss of each single bid.

We also want to draw your attention to the fact that modern SSP solutions and DSP solutions already have optimized trading algorithms based on machine learning and artificial intelligence. They work in a cookieless ecosystem that provides the best results in protecting users’ data at all levels. It is much easier to use a ready-made platform than getting into a situation where the established ad serving chain requires revision and implementation of SPO and DPO.

DPO and SPO: getting started for beginners

We have already discovered that optimization positively impacts the relationship between ad inventory sellers and buyers looking for a place to host their traffic. A skilled engineer understands that the success of advertising integrations is not about chance or luck; it is about fine-tuning and meeting user demand, improving their experience with the resource.

Next, we will examine the first steps publishers and advertisers can take to improve their engagement paths, advice provided by the IAB Technology Lab, a nonprofit research and development consortium.

Demand-path optimization (DPO) implementation for publishers

To identify media buyers, IAB TechLab recommends using a mechanism called Buyers.Json. This is one of the transparency mechanisms whose task is to:

• get rid of unacceptable or restricted content;
• avoid financial problems caused by fraudulent buyers;
• identify all media buyers to improve the ways of interacting with them further.

After implementing the Buyers.json mechanism, the next step is to reorganize your customer journeys based on the profitability of your supply chain. You can use any data for analysis, including page load time, ad conversion, invoice payment accuracy, commission paid by buyers, etc. After analyzing the information you collect, you should make a data-driven decision about reorganizing your supply chain.

Supply-path optimization (SPO) implementation for advertisers

Similarly, IAB TechLab recommends using the Sellers.Json mechanism to improve the supply-side interaction. This mechanism is used to:

• identify whether a publisher is a direct supplier or an intermediary in the chain;
• obtain data on all participants who are selling or reselling a given bid request;
• optimize further engagement with publishers to establish long-term relationships directly with authorized vendors.

We would also like to draw your attention to the primer «Understanding supply-path optimization» compiled and published by PubMatic. The idea is to use four cyclical steps that can be repeated regularly:

• assessment of what you have now;
• evaluation of engaged SSPs;
• consolidation of them according to strategy;
• and ongoing optimization based on data obtained.

This document is worth reading, and to make it more interesting, it includes tips on basic actions you can take after analyzing your data. These actions are as safe as possible, allowing you to improve the ad experience without making hard decisions. For instance, you can cut off the so-called “long-tail SSPs” that do not give you any value while taking time and resources.

Takeaways

We have considered two interconnected ways to build more transparent and effective advertising inventory bidding, corresponding to ad-path optimization (APO). The main goals are to eliminate unprofitable partnerships, increase supply transparency, and thus improve sales revenue and increase the effectiveness of ads.

Advanced platforms for running your business use optimization ML-based technologies driven by AI algorithms to achieve better results while selling and buying ads. In addition to AdTech companies, advertisers and publishers might use DPO and SPO techniques.

Demand-path optimization (DPO) is the process on behalf of publishers and SSPs, while supply-path optimization (SPO) is an advertisers-owned process that takes place via DSPs. The process runner benefits from it more, while all other parties also gain some advantages because of the improved advertisement ecosystem.

Developing and deploying optimization algorithms is crucial to the industry’s growth. The more complex AdTech becomes, the more controls must be implemented for effective financial and impression management.