The Real Cost of a Sponsored Campaign

In the past week I have had two prospective clients raise questions that ultimately stem to the "real value" of having a sponsored campaign, but more importantly, how you derive that "real value."

First of all, lets get some semantics out of the way.... Clicks are valuable; visits are valuable; reducing the CPC is valuable; brand awareness is valuable; but nothing is as valuable as a return on the bottom line. This doesn't mean just improving conversions, but improving the cost/conversion and the lifelong value of every single sale or lead. This is "real value," and sometimes being able to recognize and/or reconcile that true metric is not that easy, especially if you have a longer sales cycle and the purchase decision involves a longer than normal research phase.

So what are the issues?

If a click is somewhere in the distance past, how can you reconcile that unique visit with all the subsequent visits (whether organic, direct, or sponsored) within the research to purchase stage? What about linking this activity to a unique lead or any conversion? What about actually being able to cost all that activity to get a "real" metric.

These are difficult questions within the B2B marketplace... why?

Well for one thing, we are dealing with a fairly tech savvy market segment that doesn't accept cookies or traditional tracking methods - in fact, if anything, that market can be characterized with a certain paranoia around Internet activity and anonymity thresholds. This means it is very difficult to track click paths and activity linked to a unique visitor. Look at your analytics - do you honestly believe that your website is getting that many unique visitors? If it is, then your lead ratio is a bigger problem than you thought.

What else?

The B2B client isn't one person. In fact, there are multiple stakeholders involved in the process of research and purchase, each one visiting your website at some point; each one looking for different content and reinforcers; each one entering and exiting your site at different times, via different avenues. In the traditional method - the one that most sales departments are still built around - sales is an N2N interaction. One salesman, one client, one bottle of whatever it takes and a steak dinner (the original set for the Miller play). Traditionally, stakeholders were limited and any departmental interaction was behind the scenes, only bridged with the one presentation and office walk through. But today, there are a lot of influencers in the mix; namely, a tech influencer convincing his manager that a solution is needed, a manager looking to increase efficiencies, a CFO or CIO wanting to innovate, and a purchasing agent somewhere in there trying to dot the "i"s and cross the "t"s. Each one looking online for a solution, hopefully yours. But in reality, collectively they are one qualified lead - your analytics will tell you a different story.

Other Problems?

Throw into the mix the fact that you are not running just a simple sponsored campaign. You got smart along the way and started doing some organic optimization, usability testing, and customer research... didn't you? You are also probably trying to build a brand and have banner ads running (new research suggests that text ads are more effective if run in conjunction with banner ads - Harris Direct saw a 249% improvement in contextual ad clicks partnered with a banner buy) - maybe you are sponsoring a blog, or running unique PPC campaigns to whitepapers and internal research, or you have seen some value in site targeted content matching - either way, the funnel of your sponsored activity is wide.

Summarize

So lets summarize... you have multiple stakeholders, a long sales cycle, multiple visits, multiple sponsored campaigns, multiple entry points, multiple exit points, probably multiple products or solutions that compliment each other, multiple conversions and conversion paths, and one "real" lead. Oh yeah... and you need to quantify and cost that lead to see if your sponsored activity has "real value" - plus you have to assess a certain share of the driving force to one campaign over the other. Now for the kicker, you have analytics giving you all kinds of numbers, different numbers for different categories, and a lot of marketing hunches.

Simplify the Problem and the Solution Appears

Looking at one thing too long makes it very difficult to see anything else, when really, sometimes by changing your direction you can see something entirely different. For example, a cylinder from one perspective is a circle, from another it is a rectangle, but in truth, it is neither. So why try and solve a problem that has so many layers going into it when there is only one outcome - qualified leads.

Why not integrate your sales, at least in as far as they actually provide you with some real estimates as to how many online leads it takes to translate to a sale. Is it 100, is it 5? The average gives you a cost/lead a number you can work with and improve.

Start by working backwards, from sale to click - and if you can't get a figure without making some assumptions that you are not comfortable with - then test the assumption first. If you cannot determine if the value is in PPC or banner ads, or assess there real relationship without making an assumption that cannot be justified. Justify it! Test the outcome by turning off the PPC or Banner ads like any A/B strategy. What was the outcome? Now reverse it? What was the outcome? Can you make now make an assumption you are comfortable with?

Not comfortable with turning things off to test the outcome? Let me ask you a question then... how comfortable are you with going forward the way you have been?

It takes time, but eventually you can derive a real value for every part of your sponsored campaigns. You may even be able to reverse engineer some dynamic reports from your analytics to save you next time. The best part, you can start coming up with a strategy that lets you allocate different proportions of banners, PPC, contextual matching, etc to get even better returns - constantly testing, and constantly developing. Just remember too, the difference between play-doh and a souffle is proportions.

In addition, always keep in the forefront that you need to always define what is your absolute key KPI in the first place - that is your starting point - and any improvement to that metric is where your real value is found.

Save to: Digg | | Yahoo | Furl | Reddit