This article provides insights into how ad delivery works with TAP, describes issues that might cause delivery problems, and shows you how to deal with those issues to get your delivery back on track. The items discussed are focused primarily on live-stream delivery, but the principles are essentially the same regardless of delivery method.
What Puts Flights at Risk?
Unlike terrestrial radio, digital CPM delivery is variable and complex, which makes ad trafficking a unique audio challenge. The very things that create value — targeting and real-time dynamic delivery — can also increase the risk of under-delivery.
Broadly speaking, we can narrow things down to two primary factors that cause flight delivery problems: overbooking, and operational mistakes and client changes.
Overbooking
If you sell more inventory than you have available, then at least one flight will likely suffer. The math is simple: You can't sell 1 million impressions when you only have 500,000 available. The good news is that TAP is very sophisticated about managing delivery, and sometimes it can deal with the overlap of overbooked and non-overbooked flight dates by pushing the overbooked impressions to a later date, effectively moving the missing 500,000 impressions from this example into the future where they may be available.
Operational Mistakes and Client Changes
Sometimes mistakes happen that put flights at risk of under-delivery. For example, a flight might be scheduled to start on a Monday but the trafficker forgets to upload the creative until Wednesday. Or a flight is paused mid-flight at the request of the client but never re-activated. Sometimes the client changes their mind and requests changes after the availability has already been assessed. And of course, flight adjustments like shortening the flight end date or changing the impression goal can increase delivery risk.
Flight-level Approach to Solving Delivery Problems
TAP focuses on delivery as part of its management of your inventory and it does its best to deliver every flight at 100%. If your at-risk flights are under control, you can let TAP do its work while you focus on managing future inventory, reserving some future short-term inventory to allow TAP to "catch up."
In some cases, however, you need to proactively attack problems. This step shows you the best way to do that; by approaching the issues at the flight level and at the inventory level.
Flight-level Measures
Making flight-level changes generally requires approval from the client, but they are often the best way to manage at-risk issues because the root of all at-risk flights is that they don't have access to as much inventory as they need. So one reasonable solution is to increase the available inventory. You do this with flight changes, as described below.
Expand the geographic targeting zone
One significant way to expand available inventory is to expand the geo-targeting applied to a campaign or flight. This can be challenging with a client that has a specific physical location footprint but is sometimes acceptable for larger brands or agency-based business.
Expand time targeting
If your flight targeting includes time targeting, such as only delivering during morning commutes, this is an excellent way to assist delivery as it doesn't change the nature of the client-consumer connection. All it changes is the time the ad is delivered and the content around it. The latter point is sometimes problematic if you are selling certain content at a premium ("Being attached to our market-leading morning show is premium inventory!"), but often the time targeting is more a legacy request from spot delivery days and “dayparts,” so expanding it is a realistic option.
Add more stations/delivery types
Adding additional stations can be helpful, especially if the sale is format-driven, as format-driven sales are often based on legacy spot sales behavior. If a client never buys a station because there is an assumption that "they aren't my consumers" but the advertising is targeted to a behavioral segment, the reality is that the behavioral segment will be applicable no matter what station it is on.
You can also be creative and redirect impressions to a new flight that covers things that the client will approve, such as pre-roll or podcast delivery.
Extend the campaign/flight
Another option is to simply ask the client if they'll extend the time period of the flight. This works well with TAP's strengths, which are to manage inventory issues by moving them into the future. The danger here is that you will never be able to catch up in the future, and flights will end at some point with significant under-delivery. So approaching this at the flight level should be coordinated with a broader trafficking strategy of underselling medium-term inventory.
Change the frequency caps
There is a direct relationship between delivery and frequency caps. If a listener hears a 30-second ad and you have a two-minute frequency cap, inventory is then blocked for four 30 second units. This means that available inventory is one out of five and flight inventory is reduced by 80%. Lowering a frequency cap is one of the most effective ways to improve delivery. The challenge is user experience; hearing three insurance ads in the same break may be badly received, but this is a decision you can assess based on your own needs and expertise.
Inventory-level Approach to Solving Delivery Problems
Another approach to solving your at-risk flight issue is to look at your total inventory and to manage the ROI to improve the delivery of high value flights and sacrifice low value flights. Many traffickers mistakenly use flight-level changes to try to influence delivery in this way, but it is nearly always ineffective. The reason is that TAP is already designed to maximize delivery. Most of the methods used to improve delivery of specific flights over other flights, such as setting a flight to pace as fast as possible or increasing its priority or even its CPM, will fail as TAP natively takes over delivery management.
If you want to prioritize one flight over others, the only real solution is to pause the lower value (or low risk) flights that are competing with the higher value flight.
Once you identify the flights that are at risk and with a high ROI, you need to identify the lower ROI flights that are putting that high ROI flight at risk. Once you identify the lower ROI flights, you can then choose how to manage them.
There are a few possibilities:
Leave things the way they are if the risk of too many lower ROI flights not delivering is too great.
Pause lower ROI flights until the higher ROI flight is no longer at risk.
Some mixture of the above. For example, pausing a mid-ROI flight with a high impression goal may be better than pausing 30 lower ROI flights with low impression goals.
First, a warning: This process is complex and time-consuming. TAP is natively designed to optimize delivery with little manual intervention. As a result, managing a heavily overbooked environment to make selective changes is not part of the workflow design built into the current user interface. The best choice is not to overbook in the first place. That said, the tools are there, and this section will walk you through them.
Note the Flight Details
The first thing you need to do is note all of the delivery details for the at-risk flight. Open up the flight in TAP and note the following:
Flight end date
Subtract Total Delivered Impressions from Impression Goal
Creative length
Frequency Cap
Stations
Flight targeting rules
Enter the flight details into forecasting
Go to TAP's Inventory Availability Forecasting Report and enter the above details in the forecasting UI. Click submit and scroll down to the bottom of the page where it says Contending. Click the heading of the Contending column. This will sort the contending campaigns by highest number of contending impressions to lowest.
Choose which contending flights to pause
In the above example, the top two flights are competing for 214,637 impressions with our problematic flight. If you have 200,000 impressions at risk, you can pause these two flights to free up the inventory.
Because TAP is effective at pushing problematic flights into the future, another strategy is to sort by End, setting the earliest flight end dates at the top. Because TAP is already effective at deprioritizing flights that end in the distant future, focusing on flights that end soon and pausing them will significantly improve deliverability, as TAP would normally prioritize those flights.
The ideal way to clear inventory for high ROI flights is to find low ROI flights which end within the next 48 hours and have a high number of competing impressions. Pausing those low ROI flights will cause them to underdeliver while your high ROI flights will deliver at 100%.
A Reminder of What Not to Do
As TAP is primarily a direct-sold ad server for campaigns with impression goals, its number one priority is to deliver flights 100% to their goal. Every other component is not remotely as important to TAP as achieving 100% delivery. This means there are very few things that you can adjust in the flight UI to improve delivery.
If you have a flight ending in a few days that is pacing behind, doing things like pacing it As Fast As Possible, setting a maximum priority override, and a setting the CPM at $1 million dollars will all do nothing if the flight is competing with a flight with a 1 cent CPM that is ending in a few hours and pacing far behind. TAP's goal is be to deliver it at 100%, and nothing you can do will help prioritize another flight that is ending farther in the future.
Leveraging TAP's Strengths: Things you Can Do
Beyond the various flight level changes outlined above, you can make two major changes to the nature of the flight itself that utilize the core delivery capabilities of TAP. By now it should be clear that the two biggest influences on delivery are the flight end date and how far it is from its impression goal. This is because TAP is designed to do whatever it can to make a flight deliver to 100%. This is defined by how many impressions are yet to be delivered and how much time is available to deliver them.
These final ways to affect delivery are also the most effective for that reason:
Shorten the flight length
If a flight has 1 million impressions to deliver over the next five days, TAP will not prioritize it nearly as much as a flight that has 1 million impressions to deliver over the next 24 hours. So changing the flight end date to one closer to the current day will dramatically improve delivery.
Increase the impression goal
If a flight has 50,000 impressions left to reach its goal over the next 48 hours, TAP will not prioritize it nearly as much as a flight that has 1 million impressions left to reach its goal over the same time frame. So another effective way to significantly increase delivery is to increase the flight impression goal.
Both of the Above
Combining an impression goal increase with a shortening of the flight length will have a dramatic impact on the deliverability of a flight. Outside of flight-level changes like adding stations and geo or time targeting, there is nothing that will help you improve high ROI flight delivery more than doing both of the above.
The challenge
The obvious challenge of the above is that you are at risk of creating two differences from the original order: The flight may overdeliver if you artificially increase the flight goal (which is not always a bad thing), and the flight may end before the client wishes it to end (which is usually a bad thing). So keeping a close eye on delivery would be prudent.