Imagine a new on-demand ride-sharing company called Coolest Ride just launched in Singapore. The company wants local riders and commuters to install their Android app, and use it when they travel around the city. They know that their app comes in handy during rush hours, and provides convenience, comfort, and safety to their users.
The Coolest Ride team decides to launch a social media campaign with a goal to attract hundreds of thousands of followers and fans within three months. In addition, they also launched a print ad campaign with the biggest national newspapers in Singapore. Not only that, their marketing team negotiated to buy online ad placements on websites that are popular among Singaporeans.
But, is there a way to check if their social media followers and fans eventually installed their app?
Can they attribute a specific install to their print ad campaign efforts?
Are they 100% sure that they only served their ads to Singapore users, and not to the readers from other countries?
You’re right if you answered “NO” to all these questions, and this is where programmatic media buying comes in.
What is programmatic media buying
Essentially, programmatic buying is the automated way of purchasing online ad placements on websites, mobile apps, and other online channels. Programmatic media buying makes use of technology that leverages data and algorithms to plan, buy, and optimize ads so they are served to the right audience at the right time.
Although manually-purchased and programmatically-purchased online ads look and appear the same to the online audience, the difference lies in what happens in the backend.
Manually-purchased online ads involve negotiating with website owners directly, manually processing insertion orders, and manually placing the ads on specified ad spots for specific time periods. These ads can be seen by any visitors at any time, including those who are in other countries. The ads will also be served to both mobile and desktop audience.
In addition, the kind of audience data that these websites provide is very limited. Sometimes, they do not provide audience data at all.
With programmatic ads, the audiences’ behavior and attributes decide the types of audience that will view them. A programmatic technology accurately serves ads based on the targeting parameters specified by the advertisers.
This allows Coolest Ride to serve different types of ads to different audience segments at different times. This also makes their online advertising initiative smarter, more thoughtful, more creative, and more efficient.
How to get started with programmatic media buying
For you to get started buying media programmatically, you need to have access to a demand-side platform (DSP), like Bluagile. A DSP is a piece of programmatic technology used by advertisers, ad agencies, marketers, and media buyers to buy ad placements on any online properties.
DSPs usually have a real-time bidding (RTB) feature, which allows advertisers to place their bid on specific audience segments in real time. DSPs also connect with data management platforms (DMPs) that provide 3rd-party data and different audience segments that advertisers can easily leverage for more-focused targeting.
DSPs also connect to different ad exchanges and supply-side platforms (SSPs), like Google ADX, Smaato, Appnexus, and so many others, so you technically have access to billions of ad impressions globally at your fingertips.
Plan your campaign strategy ahead of time
Advanced DSP users and bidders may not need campaign strategies to launch advertising campaigns. Their extensive experience running thousands of campaigns give them the ability to create strategies on the fly; however, if you’re just starting out, running campaigns on behalf of a client, or working in an agency setting, you must consider putting together a simple strategy on the get go.
A well-planned strategy saves time, minimizes the guesswork during the campaign set-up and launching, avoids unnecessary spending, and keeps you organized as a media buyer and marketer.
Sample programmatic strategy for Coolest Ride
When putting together a strategy, you need to consider your advertising goal. Is it brand visibility, awareness, or conversions (in this case, installs)? If you are buying media on behalf of a client, make sure you and your client are on the same page before you craft your strategy.
In this example, let’s create a strategy for Coolest Ride with the end goal of generating installs for their Android app.
DSPs have different user interfaces, but if you are using Bluagile DSP, your campaign must always start with the general settings.
The general settings usually include the budget for the entire campaign, the types of creatives (banner ads) and creative dimensions that we plan to use, the start and end dates, our daily budget, and the number of times we want our ads to be shown to a unique user (frequency cap).
Budget: Let’s have a month-long campaign with a budget of $15,000. We will limit our daily ad spend to $500. This means that once the campaign has reached $500 daily spend, it will automatically stop serving to the audience for that day and will restart again the next day. This way, we won’t spend all our budget in just a few days.
Ad units: For the ad units, we will focus on popular mobile app ad dimensions: 300×50, 320×50, and 300×250. You are free to use other dimensions if you like.
Landing URL: The landing URL is where the user directs to after they click on the ads. In this case, we want to direct the user to the Coolest Ride Android page on Google Play.
Frequency cap: We will set the overall frequency cap to 3 times every 24 hours. We do not want to bombard and annoy the audience with so many ads in one day.
Sample Strategy – Android App
Under the main campaign settings, we can create multiple strategy settings. For this campaign, however, we will only create one sample strategy.
In this strategy, we will use the same ad units, the same the landing page, the same start and end date, the same frequency cap, and the same target inventory type (mobile apps).
We are setting the overall budget to $7,500, with a daily budget of $250. You are free to set whatever budget that you want as you see fit. There are no rules.
Supply sources: The supply sources are where the audience / online traffic will come from. Most DSPs connect with multiple ad exchanges, so you can select specific ad exchanges, like Google, Pubmatic, OpenX., if you are familiar with the quality of their traffic audience. In this case, we will target all open sources / ad exchanges available.
Location: Our target location is Singapore because we only want our ads to be shown to Singapore audience.
Goal type: Our ultimate goal is app install, so we’ll put CPI (cost per install) here.
Maximum CPM bid: This is the amount that we are willing to pay per 1,000 ad views from our target audience. In this case, we are willing to pay a maximum price of $1.75 per 1,000 views. Since most DSPs work in a real-time bidding environment, we may be charged lower than $1.75 depending on the competition at any given time, but we won’t be charged higher than $1.75 for sure.
Budget pacing: We have the option to spend our budget right away, or pace it evenly throughout the day.
Day part: Day parting means the hours and the days that we want our campaigns to run. For now, we will let the campaign run 24/7.
Device: We will target mobile users only.
Operating system (OS): We will target mobile users using Android OS only. If the Android app is compatible with Android 4,5,6 only, then you should target those OS versions only.
Browser: The type of browser that the users use is irrelevant in this campaign because we are buying placements on mobile apps, not on mobile responsive sites.
Connection type: It is important that we target users with mobile data only because that’s needed to use the Coolest Ride app, especially when the user is out and about or stuck in the city. In this case, we will target 3G and 4G users only
Carrier / ISP: It’s irrelevant in this campaign, so we will leave it blank.
Now we have a highly-targeted audience segment for our Android app campaign: Singapore Android 4G and 3G users. This audience segment will more likely respond to the ads and install the app compared to the general Internet users and newspaper readers
Execute the strategy using a DSP
So, now, let’s use Bluagile DSP to execute this strategy. You can follow the step-by-step guide on this page.
Run a test campaign for a limited period
Instead of going full force right away, we will observe our campaign for a limited period first, and maybe with a smaller daily ad spend, like $50-$100, which we can edit in our campaign settings. This way, we get a glimpse of the types of data and results that we expect to generate.
We also need to do some optimization to refine our campaign further. Some advertisers check if they can already realize some ROI during the test period, and then decide if they should scale the campaign or not. They also want their ads to appear only on brand-safe publishers, and make sure they do not get bot traffic.
Identify insights from the initial reports
During our test run, we will focus on the best-performing publishers (mobile apps that served our ads), as well as the best-performing creatives. We will also check the time of the day or week that majority of the conversions / engagements occured. Is it from 7pm to 11pm, or is it during the morning? We will also check the best-performing Android OS versions.
Once we already generated new insights, we’ll create another strategy and apply these insights. For example, if most of our engagements and conversions come from 20 mobile apps only, we will target those by creating a “white list” containing the names of these mobile apps. If one creative is performing better than the rest, we will use the best-performing creative in our new strategy.
The optimization process should be continuous and should be done on a weekly or bi-monthly basis, until we get the results that we want. Creative and analytical thinking, plus relentless campaign optimization are keys to successfully drive installs and generate more users in any mobile app campaigns.