How Do We Go From “Risky” To “Reliable”?

How Do We Go From “Risky” To “Reliable”?

A couple of related posts caught my eye this week. Network operator Raji Kaira penned (I suppose no one really “pens” anymore, but it is still sounds good) a guest column on Sixteen:Nine lamenting the impact of dead screens on the perception of his agency friends. The ad buyers told him that seeing dead screens made them view digital signage in general as a risky buy, in that they can’t be sure that promises are kept. That’s a pretty serious issue, as Raji points out. Stephen Ghigliotty of ScreenVox took the conversation a bit further in a blog post entitled “Compliance: The Cost of Doing DOOH Business”. Stephen correctly pointed out the absolute necessity of knowing what media has actually played correctly, as well as a few hardware and software checklist items to be mindful of. Without doubt, if we as an industry want to get the full benefit of the increasing number of positive studies with respect to advertising effectiveness, then we have to minimize damage done by non-compliant networks and screens.

Here’s the dirty little secret: No network of any scale will ever be 100% compliant, and that is not a black eye for digital signage. The endpoints in any Neilsen or Arbitron rating study of broadcast/cable TV are not 100% compliant, either, and this is a known fact. Many are on but unwatched, and many have the sound turned off,. That does not detract from the credibility of the ratings numbers, nor should it. It is accepted because it is understood. What we need is similar understanding.

We’ve posted before on compliance, but what I’d like to examine today is why screens become non-compliant, what reasonable expectations of overall compliance might be, and how networks can gain credibility by managing and positioning compliance.

Delivering a piece of content to a screen at a given time requires many things to go right. There are many points of failure between the content distribution server and the display. Each of them can impact playback and compliance.  Let’s list a few:

  1. The internet: The vast majority of digital signage networks are dependent upon the internet to receive content and playlists and to send player logs. An interruption of service could impact the timely, complete delivery of content to a player. A slow or intermittent connection could have an equally negative effect.
  2. The network and network connection: The head end of a premises network usually consists of a modem and a router and/or switch of varying complexity. The connection to the media player is either wired or wireless. Proper configuration and operation of the modem and router are of course important and usually extremely reliable. In a wired ethernet connection, the physical connection can be inadvertently or intentionally (see #8 below) disconnected, causing an outage. In a wireless environment changes in the environment, including physical changes that impact radio signals and configuration and security changes, can impact connectivity and therefore compliant playout.
  3. The media player: As Stephen’s post noted, the media player is the hub of all digital signage activity. It is a computer, an electronic device. It basically does what it is told, but there are many things that can cause a hiccup. Corrupted files or power spikes can cause glitches or reboots. Software errors can do the same. A well-designed player will recover gracefully from errors, and will have the ability to be managed remotely when technical intervention is required.
  4. The power supply: Often overlooked, the actual flow of reliable, steady power is critical to the proper function of electronics, particularly media players. Players that are on the same circuit as walk-in coolers, air conditioners and ovens are subjected to frequent spikes that can degrade performance. Brown-outs, temporary outages and grid-related spikes can also damage players or force reboots. Not many owners want to step up to power conditioners or proper UPS devices, but they extend player life and minimize outages.
  5. The software: Software is never completed. If you doubt this, think back on how many times your desktop apps, browser, operating system or smartphone has been updated in the past 3 months. Bugs and general anomalies can and will happen, despite best efforts. Most are easily fixed once identified. All of them can impact playout, of course. Properly testing your software of choice and understanding how your provider manages version control, patches and upgrades can help put your mind at ease.
  6. The video connection: Simple, but important. The cable connection from the media player to the display is a potential point of failure. Yes, there are occasionally bad cables. More often there are improperly connected or disconnected cables. No signal, no picture.
  7. The display; Displays are electronic devices, too. As such they are subject to all of the bad things that can happen to the player. They also get hot (as some players do), and improper ventilation may cause malfunction or failure if heat builds up beyond specified tolerance. Displays can also get reprogrammed to the wrong input by anyone with access to a remote or the screen control buttons.
  8. The people: Lest we forget, people come in to play here. People deploy and connect the devices. People develop and launch the campaigns to the network. People may choose to manually change volume or turn screens off. People may pull a plug or a cable by mistake or intentionally. People may not care enough to report a problem. People are the hardest part of this chain to manage,  and unfortunately, I have yet to meet one with a handy RS-232 port.

Even with all of these points of potential failure and non-compliance, technology exists that gives us the ability to both monitor and manage all aspects of performance. It also gives us the ability to understand whether a screen was functioning as desired when a piece of media was played out by the player. Properly specified and deployed components, good software and well-trained people provide for a network that can be expected to operate properly (i.e., be compliant) at a rate well above 95%. While all good operators strive for 100% compliance and are never happy with anything less, there are enough gremlins and gotchas out there to reasonably expect a certain number of non-compliant screens at any given point in time. Like the unwatched or muted TVs, the non-compliant screens are a fact of life. To be sure, causes must be identified and addressed, but the pursuit of perfection is likely to be recast as the pursuit of realistic optimization. Part of optimization is actually knowing what elements are sub-optimal.

Given all of that, savvy operators do not base their rates or performance metrics by extrapolating out a 100% compliant playout forecast. Instead they sell on a number that accounts for likely compliance. That way, if an anomaly occurs during a campaign, there can be enough slack and time to make good on the promise. Similarly, if everything works swimmingly, the advertiser will likely get more than they expected, which is good for business relationships.

Raji is correct to raise the issue of the ripple effect of dead screens. Stephen is right to remind us that compliance is the price of poker in an ad-supported network. But the sooner we help our ad buying partners to understand that we are not any more perfectly compliant than a home viewer of broadcast TV; that we are on top of compliance monitoring; that we relentlessly optimize performance; and that we sell based upon an expected net compliance percentage, the sooner we will be seen less a “risky” and more as “reliable”.


  1. Dave Haynes August 26, 2010 at 6:47 pm - Reply

    Good perspective. I would actually put people at the top of the list if we’re ranking the most likely points of failure.

    Anybody starting to deploy a network who thinks the people at the venue are going to be field support are in for some major forehead-meets-desk sessions.

    Where I disagree a little is with the Shit Happens thing, and how media people know that and factor it in.

    Yes, true … but when I actually emerge from my home office lair and go out into the real world of malls and stores and bars and restaurants, dead screens are not a rare sighting. They’re a regular sighting.

    I’d suggest planners and brands see too much failed stuff in the field to just write all that carnage off as anomalies.

    PCs and other devices do fail, but there are many, many, many ways to limit the risk of failure and quickly remedy those failures when they happen.

  2. Ken Goldberg August 26, 2010 at 6:59 pm - Reply

    Good point on not merely writing off dead screens to anomalies. Sadly, too many of the networks out there are not capable of remotely managing their screens and players, or have a plan for addressing issues when they arise. “Dead” screens should have a “heartbeating” player attached that can be accessed to diagnose and usually correct whatever caused the outage. If you wait for someone to call, it may be a long wait, as you also point out!

Leave A Comment