This will be the last of the "Calculating Digital Signage ROI" articles for a while.  However, due to high demand, we're probably going to be augmenting and re-packaging the whole series as a downloadable whitepaper in the near future, so if I've missed anything that you'd like to hear about, let me know and I'll try to add it in before then.

Before we begin, I'd like to apologize for the relative dearth of posts lately.  WireSpring has just moved to a much larger facility in Fort Lauderdale, and I've had to switch from blog-writing mode into logistics mode for the last few weeks.  Now that things have settled down, I expect to get back into the regular swing of things.  

Also, if you've missed any of the other articles in this series, you might want to go back and read them first:

1: Calculating Digital Signage ROI: The Ground Rules
2: Calculating Digital Signage ROI: Understanding the Limits of Your Data
3: Calculating Digital Signage ROI: 3 Metrics that Matter
4: Calculating Digital Signage ROI: Methods to Gather your Data

By now, hopefully you've been able to figure out a way to acquire the necessary data to put together your digital signage ROI analysis.  Depending on how large of a network you have (in terms of both number of screens and amount of content), how long it has been running, and of course, your ability to execute, your ROI estimate may be great, terrible, or somewhere in between.  Like everything else in the world, the success of your deployment is going to be largely based on the time, energy and resources that you put into it.  The more you give, the more likely it is to be successful.  However, even the best laid plans and the most well-capitalized projects can provide lackluster results, or worse, fail altogether.  As a digital signage project manager, you will need to learn how to identify an ailing installation and take steps to improve its performance while you can.  I'd like to bring up a couple of ideas to keep in the back of your mind while analyzing viewership or purchasing metrics, in pursuit of determining the ROI of your digital signage network.

1. Set goals ahead of time
This seems so obvious that I don't understand why more people don't do it.  Make a good estimate of what your signage network is capable of, and then make that your aim.  If your goal is to divert foot traffic to a specific area of a department store, or a particular store in a shopping mall, your content and installation is going to look different than if your goal is to boost sales of a loss-leader product.

2. Give it a chance first
Once you've spent a huge amount of time, energy and money deploying a network - even a small one - it can be frustrating to have to wait to see how it works.  But one of the most common pitfalls of a digital signage project is to judge it too early and start making tweaks before the system has had a chance to settle down.  It goes back to the whole notion of correlation and causation that I discussed in a previous article (I believe there was an example mentioning tube socks).  Without enough baseline data for how your system works, it's hard to decide when a modification to placement, content or scheduling needs to be made.  So give each installation a few months to settle down, and then observe the impact it has on traffic flow and store performance before making changes.

3. Make a project plan, or at least a timeline
Put together a reasonable timeline for achieving each goal that you've specified.  For example, if your overall goal is to boost sales of advertised products by 10%, make an estimate of how long that should take, how much leeway you're willing to give it, and what sub-goals you hope to achieve along the way.  Also, decide who in your organization is going to be responsible for reaching each goal and tracking the results of the system. By specifying these things ahead of time, you'll have some measure of accountability (even if it's just to yourself).

4. When it comes time to make changes, take baby steps
There will come some point in time when you've maximized the performance of your digital signage network, and will need to make either structural or procedural changes to further improve results.  For example, you might find that you need to add or take away screens, move them to a different floor height or location, or change your content and scheduling.  While it might seem like a good idea to get all of your changes done at once, our customers have found time and again that it pays to test!  Move from a 15 second spot to a 30 second spot and test the results.  Add endcap displays to your ceiling-mounted network and check to see if there is actually any benefit.  Our motto at WireSpring is "test early, test often" and that should apply to more than just software.  Every aspect of your network should be both subjectively and objectively tested, not just during installation, but throughout its lifetime.

Finally, I was debating adding a section on "explaining away anomalous numbers" or "manipulating statistics to your advantage," but my ethical side got the better of me :)  In all seriousness, statistical analysis is a subject that has filled entire libraries, so I don't think I could really do it justice here.

By this point I hope I've shed at least a little light on how you might go about building an ROI analysis for a narrowcasting network.  While it's certainly not a complete guide, everything I've mentioned has proven useful either to us or to our customers, and during a brief Google search of the web, I wasn't able to find a single repository that has listed most of these ideas in the same place.  So while I certainly can't take credit for inventing everything presented here, I do hope that the collection of these ideas will inspire you to sharpen your creative and empirical skills, gather your data, and really determine the value of your network once and for all.  If your results aren't what you hoped them to be, perhaps these pages will give you some ideas about how to improve things.  And if you're at or above your desired targets... well, there's always room for improvement!