On our website, we define programmatic advertising as follows:
Let’s first take a look into each of these three important aspects of programmatic advertising.
First, an intro.
It’s easiest to think about it like the stock market. There is a buy side, a sell side, and exchanges in the middle that connect both parties. Buy sides major players are DSPs, or media buying software that trading desks and ad agencies use to purchase programmatic media. The sell side major players are SSPs, software that allows publishers and content owners to offer inventory to exchanges to be bought by DSPs. As you get deeper into programmatic advertising, you’ll learn about all the other acronyms, and the other supporting players from the LUMAscape. For right now, we think knowing the above is a great start.
The Lumascape details the adtech industry and its major players.1
To begin, we just want to say that we highly recommend the below training from the Edge Academy from The Trade Desk if you’re serious about nailing these concepts. We’ll do our best to summarize and briefly explain the technical side of how programmatic advertising buying works.
Marketing Foundations Edge Academy
Now we’ll get into the specifics of programmatic. Programmatic advertising is often referred to as RTB (real-time bidding). Instead of buying ads via an I/O, or in another up-front manner, programmatic advertising looks at the metadata in each impression as a page is loading (<100ms) and determines if it’s worth bidding on. If your ad is the highest bidder, you win and your ad is shown. As you can imagine, this can help you make sure that your advertising budget is data-driven–buying impressions only when pre-determined requirements and users are matched.
(Technically, there are other buying methods for programmatic besides the real-time, auction-based buying method, but RTB is the most common, which is why we’re using the terms interchangeably, despite there being small nuances with each term.)
Programmatic advertising is transacted via software platforms known as Demand Side Platforms, or DSPs. DSPs vary in capabilities, effectiveness, and cost. Choosing a good DSP depends on your needs, budget size, and strategies. Not only is it important to choose a DSP that fits your business, but it is just as important to choose an agency that has experience with your preferred DSP. Megagon Ventures helps you check all these boxes. We partner with StackAdapt, TheTradeDesk, Basis, and Simpli.fi. We feel that these DSPs offer the greatest coverage for brands looking for objectivity, transparency, and results-driven advertising.
Programmatic advertising is extremely efficient. When you make the switch to programmatic advertising, you enter the world of getting better, measurable results, without the waste of traditional advertising. Waste is eliminated on two fronts. First, you have complete control of your targeting model, which determines exactly when and when not impressions are bought. Second, you are able to measure every single impression. So much of traditional advertising methods aren’t measurable. You can’t measure how many people saw and were influenced by your billboard or magazine ad. But with programmatic, you can see exactly how effective your ads are.
We have another article that covers the benefits of programmatic, but to summarize, programmatic is optimal for eight reasons.
We’ve already touched on this above, but programmatic eliminates wasting your ad budget on low-performing ad impressions.
Programmatic is objective. You should make sure that your DSP only operates on the buy-side, and doesn’t own an exchange or their own inventory. Objective DSPs like TheTradeDesk guarantee that your interests as an advertiser are always represented–even ahead of their own profits.
Programmatic advertising is measurable and attributable. As we stated above, you can measure each impression’s impact on your KPI, and even attribute users and impressions that led to a certain event like a click, view, or conversion.
Programmatic advertising is transparent. Throughout history, advertising has been a black box of mystery. You spent money on it, but never knew exactly what was working. Programmatic advertising gives you insight into what is and isn’t working, and the control to modify your data-driven strategy in real-time.
Programmatic advertising takes targeting to another level. You can target with your first-party data, or with any number of third-party data segments built into most DSP’s data management platforms. You can target based on a web page’s content, or you could target based on the weather. You can target people by their interests or demographics. The possibilities are literally endless.
Programmatic advertising is an expansive universe of channels, mediums and inventory. We’ll hit some of this below, but in short, programmatic is so much more than display ads. You can target music and audio, video, mobile apps, native ads, connected TV, digital out of home ads like in airports or connected billboards. Within each of those channels or mediums, the inventory you have access to is mind-blowing. But any website, app, TV show, or place with ads is or will be able to be bought via your DSP. And with the growth of AVOD, and Netflix’s recent announcement to add ads, programmatic’s future is bright.
Programmatic advertising is flexible. You can immediately turn off budgets and strategies that aren’t working, instead of waiting for the post-campaign wrap-up. You can increase or shift budgets between strategies that are working. You aren’t locked into spending money where it’s ineffective.
Programmatic advertising is data-driven. Data is the heart of programmatic advertising. It influences your planning, execution, and learning. So much data is available, powerful machine learning models help you get additional insight, and skilled programmatic media buyers can help you dive deep into and derive insights from the vast amount of data.
Next, we’ll take a journey into planning, execution, and reporting on a campaign to help you learn even more about what programmatic advertising can do.
This is how we normally go about planning, executing, and reporting on campaigns. This is a super high-level overview, and is just to help you better understand the process. We have supplementary articles explaining each of these concepts in detail.
Planning. First, it’s important to know any business objectives. That helps align your advertising objectives with the business, which ensures greater success on your advertising campaigns. Once those are confirmed, the next step is to set specific, measurable advertising goals, or KPIs. In this step things like funnel-placement, creative strategy, inventory, messaging, targeting, etc. are solidified. When using a DSP like The Trade Desk, you can take advantage of powerful machine learning and artificial intelligence to help you plan and forecast as well as get helpful feedback.
Data and Targeting. After we have a concrete plan, we can then move on to targeting. Most DSPs are equipped with DMPs, or data management platforms, that you can use to help you reach a certain audience. These DMPs allow you to upload your own first-party data, take advantage of third-party data segments from data providers, create retargeting strategies, develop lookalike models with the power of AI, and even do things like target interests. Other targeting strategies like contextual targeting, weather targeting, time of day, inventory, etc. can be added during this step. However, regardless of whether you add it up front, you can always modify each targeting strategy once the campaign is live based on the data and the effectiveness of the strategy.
Channels. Based on whether your campaign is an upper, mid, or lower funnel campaign, you may want to utilize different channels, or mediums, for your advertising campaign. Each of these channels has different strengths, and it is important to also consider your messaging along with your objectives as you determine which channels you want to advertise on. The best DSPs nowadays are omnichannel, meaning that you can advertise on every channel in the same platform. Some of these channels include:
Creatives. Once you know your channels, you can then work on your creatives, including assets, messaging, colors, and content. The type of creatives you will need to develop are determined by the channels you’re targeting. To keep it simple for right now, remember the two following points. There are four main types of creative: display (banner), video, audio, and native. You can have hosted creatives or third party ad tags. Creatives are usually provided by the advertising brand, or can be created with the help of a creative strategist.
Tracking. When users click on an ad, they are directed to someplace, usually the brand’s website. It is important that you are able to track the people that land on your page for measurement, attribution, and for retargeting purposes. This is usually done by placing a small piece of code generated from the DSP into the head of your website’s HTML code.
Inventory. You have a variety of options when deciding the inventory you want to target. You can target all site, utilizing DSP block lists to block sketchy and low-performing sites, and then further optimize once the campaign is live. Alternatively, you could choose to target only specific sites, apps, or publishers, and by default block the rest. In addition to these options, you can choose to target private marketplace deals in the DSP’s built-in PMP library. These deals are premium inventory and ad slots, but their higher CPMs are offset by their premium placements and high results.
Measurement and Attribution. Finally, if you haven’t done so already, you’ll want to make sure that you know how you’re going to measure the effectiveness of your campaigns. This becomes especially important when your KPIs are conversion-based. For example, if multiple ads on multiple platforms are claiming credit for the conversion, how do you decide which one it really was? There are numerous methodologies such as last-click/view, first-click/view, time-delayed, linear, position-based, and data-driven attribution. Each one has pros and cons. We go into more detail on this subject in another article.
Execution and Optimization. Finally! the fun part is here. With all your settings confirmed, you’re now ready to start buying media. Media buyers (traders) carefully watch each campaign to ensure that it gets off to a good start, and then starts pulling reports to optimize the campaign to meet its goals. Again, we don’t go into all the detail here, but you can optimize almost anything. What you do optimize is primarily determined by your KPIs. For example, a click-based KPI means you could adjust your bidding CPM upwards $0.50 when bidding on inventory that is above the fold, and downwards $0.50 when the ads are below the fold. You could also bulk adjust your bids based on past site performance, browser and device characteristics, time of day and day of week, and the frequency that the ad has been seen by the user, just to name a few. Continual, consistent optimizations result in optimized campaigns that meet KPIs, minimize waste, and drive real results.
Reporting. Throughout a campaign your agency should provide you with reports so that you can understand the progress of your live campaigns. At the end of the campaign, you are presented with in-depth Excel sheets and PowerPoint slides that highlights the insights gleaned, with emphases on what worked and didn’t so that you can use those learning as the input into future campaigns.
Wow, that was a lot. And it was the simplified version. Just grasping these concepts alone should provide a great high-level understanding of programmatic advertising, it’s benefits, and the process by which a campaign is executed. We recommend that you study this, and view our other supplemental articles to increase your understanding in each specific area. You’re well on your way to becoming a Programmatic Master!
