At various points in my career, I’ve worked with various marketers who all dream of closed loop marketing, where they can account for the direct impact of their efforts on sales. If sales is like an art and science and marketing is much the same way. These days, everyone wants it to be more like the latter, with more predictable and demonstrable results.

In a previous role, I worked at a demand generation software company that helped companies market their wares and track how effective their campaigns were. Besides email and microsite campaigns, they had a pretty cool feature that tracked a user’s behavior on a website. What separates this from your typical analytics program, was its ability to provide very detailed tracking on a per-user level and if the user clicked through to your site via email, we’d even know exactly who it was, how often they visited the site and what they did there. It was the online equivalent of keeping an eye on people in your retail store – what products they picked up, what labels they were reading, which sales people they spoke with and what items they were thinking of buying. The software had the ability to aggregate all of these activities automatically and rate the quality of these web-based leads depending on what you deemed important. For instance, if a user came to a site through an email campaign, clicked through to the site but then left, that person would rank pretty low. But if another user clicked through, visited a few times, downloaded a whitepaper and signed up for a webinar or demonstration, that person would rank quite highly. The lead scoring program would notify sales people who performed a certain number of these valuable activities and if the user did enough of them, he or she would be considered a lead and thus worthy of a follow up email or invitation to an event. For companies that want to allocate their sales resources strategically, this is a great time saver because it helps separate the tire kickers from the true spenders. It also gives insight into which companies might be interested in your own company.

More recently, I’ve been looking into linking behavior with business results. Web analytics has certainly come a long way from the days of downloading huge log files and parsing them into your desktop software that made pretty generic charts and graphs. I would say the capabilities of companies like Omniture, Webtrends and Google Analytics has come a long way since then. Our ability to track activity and visualize data is amazing. We can even understand which web pages are putting in the most contribution to conversion, a term I first learned about in accounting class. But when it comes to inferring user intent, the picture becomes less clear. For instance, I can know what % of visitors ultimately end up converting, but what are the activities that are required to make that leap? And what if the leap requires visiting other sites that you can’t track or analyze?

That’s the next step that so far, few if any companies have been able to take. Here are a couple of scenarios:

• A site visitor is about to sign up for your service or buy something online but wants to read a third party review before committing. They leave, visit any number of sites from Amazon, Yelp, Epinions etc. and then sometimes come back to buy.

• Someone visits your site via search and decides to spend time there. Next, the user searches for your brand to get near real-time commentary on Twitter before coming back to sign up for a trial.

See what’s happening? Just like in real life, people are increasingly gathering information elsewhere particularly on 3rd party sites. But for a marketer, how do you quantify the value of these interactions not happening on your site? How do you score leads happening out in the blogosphere, forums or Twitter? As we got closer to closing the loop, it opens up and gets even broader and even more complex than before. So what’s a marketer to do? Ideally, there would be a super tracking cookie that could send back data on sites users visit, but privacy concerns don’t make it feasible. Here’s some general thoughts on alternatives:

• Monitor the social space and get a pulse on the quantity, tone and influence key sites/destinations have on your customers. There are tons of tools that can help you do this, including Radian6, TruCast, Cymfony and others.
• Get a baseline understanding of your brand’s chatter and draw correlations between volume and tone of discussion with product releases or announcements.
• Segment this incoming traffic from social sites and compare with direct site traffic, search engine and campaign traffic. This will give you an idea of what impact the social sites have in comparison. It’s not perfect, but it’s a start and right now, it’s more art than science.
• Keep at it. Marketing is a constant evolution, so the more you can observe, the better you can refine and the better you can deliver to your customers.

A recent post on Mashable caught my attention, a how-to guide on measuring online influence. It begins with a dictionary-type definition of influence followed by the requisite discussion of personal branding and what it means online. Where things get interesting is how the author dives into mathematical formulae of how to measure influence with the following equations:

Influence = (Personal Brand * Trust * Expertise)

Of course, since Expertise = (Knowledge * Trust), we can further refine the equation to:

Influence = (Personal Brand * Knowledge * Trust2)

I am all for quantifying results but how does one come to these conclusions and more importantly, how does one make these calculations? Whatever happened to integers and real values? I will agree that influence is made up of these components, but to stating it in these terms isn’t helpful nor is it actionable. Some units would be a start.

The next part of the article delves into measuring influence by looking at tings like traffic, connections, track record and so forth. All interesting and valuable metrics. But where the article falls short is defining how these are related  and what to do with the metrics. They are data points without a story to wrap them up in a logical way. They’re sort of like ingredients without a recipe; they’re all important but without knowing what to do with them, they’re just ingredients. I admire the author for putting these ideas together but at the same time, let’s be careful not to position it as a how-to guide when the step-by-step part of it is missing.

I mentioned in an earlier post that quantifying things with rigor is the way to go and in this economy, clients are increasingly expecting this. If it were easy to do, it would have already been done. There are no clearcut accepted ways of doing this yet, so I’m all ears for looking at new methodologies and approaches. Some will work, others won’t. But shying away from facing reality isn’t going to cut the mustard these days. This is something we will be hearing more and more of, so I’ll be revisiting this topic in the future.