DSPs, SSPs, DMPs, Ad Exchanges… the list goes on. The programmatic landscape is filled with acronyms. It constitutes a complicated and somewhat confusing ecosystem that ties together the key players in the industry; advertisers and publishers.
As media buyers on behalf of our clients at Merkle, we spend a lot of our time using a DSP (Demand Side Platform) to set up and manage client campaigns. The DSP is essentially a platform that we use to connect to Ad Exchanges (large aggregators of publisher inventory) in order to buy advertising space (inventory) on the internet and manage how we execute this buying of inventory in real-time.
When it comes to deciding which DSP is right for you and your company, there are a range of factors to consider. How user friendly is the interface? What are the core reporting capabilities? What kind of audiences can you target? How easy is it to make large automated changes? How well integrated is the DSP to access CRM data? What kind of brand safety can be applied? This list really does go on and the answers to these questions will likely change over time, dependent on your business needs and campaign strategy.
OK, what’s out there?
Whilst you may have heard of the Google Marketing Platform (GMP), which encompasses a whole suite of products including DV360 (Google’s DSP), there are a wide range of other DSPs available, each with its own level of maturity, benefits and challenges. This blog will explore the capabilities and nuances of the AppNexus Programmable Platform (APP) and how it may be able to benefit your programmatic campaigns.
What is AppNexus?
Founded in 2007, AppNexus was the first specifically Programmable DSP available to marketers. After significant investment, APP was launched in 2017, offering greater control and transparency to the media buyer, as well as the opportunity to fully harness internal programming talent.
So, what’s all the fuss about?
Splits have a range of uses that enable the campaign manager (us) to be more efficient, manage multiple different KPIs simultaneously and better split out targeting strategies to deliver stellar performance for our clients.
This is done through splitting up the inventory that we are targeting with specific budgets, goals and bids per condition. An example of these conditions could be a city or a specific website. We then place a certain value/bid and budget allocation on each of these conditions as well as decide which creative and devices we want to deliver this on. The APP will then work to optimise towards these goals and the value we place on each condition, constantly learning from each bid made. It is undoubtedly a quicker way to optimise performance and allows you to granularly target specific inventory, identify the high performers and constantly test to tease out further conditions that are driving your performance.
Like other DSPs, APP uses algorithms to help optimise towards your performance goal. However, in this case you can set a specific performance goal and program the DSP to optimise towards that goal. For example, CTR, CPA or CPC. There are 2 stages to this process. Firstly, the Learn Phase. This is where the APP will collect and analyse data on all the impressions that you have won. After it hits a level of statistical significance you will then reach the Optimisation Phase. Here, the APP will constantly vary your bids based on past performance of similar inventory. Therefore, from building a picture of what specific domains and targeting work well, it will bid more on these valuable impressions, with a higher likelihood of hitting your conversion goal.
APP has an open API. This essentially means that the platform is an open book in the way of automating and creating custom models to run your activity. If you have a strong data science team, this is definitely worth looking into. It also means that you can develop and integrate scripts to automate otherwise manual processes and alleviate workload so that you can focus more on strategy.
Console Integrations with 3rd Party Tools
Unlike other DSPs APP has some really useful third-party products which are directly integrated into the platform. Think of them like an extension on your web browser, but for a marketing platform. With Grapeshot for example, you can easily build out keyword targeting as well as brand safety segments to exclude. This ensures that you are only appearing on brand safe, quality inventory, when and where you want to.
Is it for me?
Ultimately, APP is a well-integrated DSP offering some stand out optimisation capabilities. Whilst the interface is arguably less user friendly than other platforms, the true value can be realised through fully utilising the open API.
So, if you have a great team of data scientists, who are keen to develop their own algorithms to enhance your bidding strategy, this one is for you.
For any questions on which DSP is right for your brand, feel free to reach out to the team.