Measuring marketing impact across omnichannel is a huge challenge and a pain point for today’s retailers. Systems and media tracking don’t “connect.” The customer journey and engagement across touch points is also highly situational – it varies a lot by category and price point. Customer patterns vary a lot by season – stressed holiday shoppers may purchase more online. There is NO easy solution for crediting attribution of which engagement or touch point accounted for the actual purchase decision. While it is possible to track touches within media, it is very difficult to track transfers across media and platforms, especially digital. The whole is greater than the sum of the parts. The best case for retailers is to systematically test different media mixes and offers across media and channels, and then measure impact on both traffic and conversion. Very similar to A/B testing in email campaigns, but with a more macro view of multimedia and mix in the tests. Retailer’s do NOT have to be able to measure the “causality” in order to derive value.… Read more »

It’s vital for retailers to measure the impact and results of their marketing programs. In many cases the process is very complicated, particularly when measuring the impact of brand awareness and community-building campaigns.

Ultimately, the most important indicator of a marketing program’s success is increased sales. Therefore, programs that are directly linked into the retailer’s POS system are likely to be the most impactful and measurable. Retailers should use promotions and offers that can be activated and redeemed in their POS system.

This pain point has been recognized as a pain point for decades. It is extremely complicated and maybe impossible. While it may be possible to measure exposure to advertising campaigns, it’s not possible to measure engagement with the advertising and the incremental influence of other communication such as PR or visits to a store or websites that showcase products (but not necessarily the retailer). For the most part retailers examine their media spend overall and see how it correlates with sales. There are algorithms in media that help distinguish contributions of various channels overall. However, getting those algorithms to be used effectively, i.e. profitably, for the marketer could be elusive. Finally, what is least often addressed (if addressed at all) is the impact of execution and content. I don’t think a marketer should implement or dismiss a media plan or its components without knowing that the campaign itself has merit. And if the advertising shows promise in one channel, it doesn’t mean it will be effective in another channel or, as the article suggests, with… Read more »

Attribution is THE obstacle to delivering on “omnichannel.”

The reality is each sales channel (web/phone/store) can support each of the others. The problem is they are typically treated as separate P&Ls and therefore have very little incentive to drive omnichannel behavior.

The first part of the question proposed in this discussion is, how do we determine the need for measuring cross channel effectiveness? From this information we will know if the process is necessary and, if it is, what parameters must be addressed for measuring and reporting. In this economy, or any other for that matter, spending money to maintain efforts that produce little or no results is unacceptable. We should also be prepared to assess whether or not the proposed cuts aren’t in need of repair. Over-cutting marketing and advertising makes up a large portion of the harmful decisions executives make today. Much of the separation from market effectiveness we see today is due to executives replacing plan development, refinement and execution with an ad hoc, with eyes crossed, “vision for the company.”

There is no easy process and there are no easy answers on cross-channel attribution. Or rather, if anyone is giving you and easy answer, you ought to be highly skeptical — the analysis is probably wrong.

Testing is the right answer for understanding causal impact of marketing investments in a noisy world. But you have to do testing right and that isn’t easy. First, you have to be measuring actual sales and profit data. Second, you have to have the right control for the test. Third, you have to avoid the pitfalls of simply tracking exposed individuals. Fourth, you need to do much more of it.

Marketing to customers is multi-touch or, in other words, continuous engagement. Which of the many touches creates the final purchase decision/action is anyone’s guess, and guess is about as good as it gets. Marketing-to-sales success is probably as good a measurement as one can have. The other measurement would be for marketing to say, “how about we just stop doing all of this and do nothing instead?” and see the reaction.

It is definitely a challenge to measure the effectiveness of cross channel campaigns and even more difficult to attribute sales to a specific channel touch point. Recency Frequency and Monetary Value (RFM) is not complicated to implement but tracking the specific touchpoint(s) attribution is more complicated. We have used Salesforce.com effectively for attribution but few retailers have implemented Salesforce as a business to consumer solution as most of their efforts have been business to business. Maybe we shouldn’t be even trying to attribute sales to a specific touch point like the “last-touch” model, as the consumer’s decision process is often impacted by many brand interactions — not just the last one. A better approach is to look at it from a holistic perspective. For example, measuring the sales impact from marketing campaigns that leverage different combinations of multi-channel communications. What combination and order of marketing communications are most effective? Also, based on this analysis, we can develop a better understanding of the customer journey across channels. Oh, and by the way, the Holy Grail is to… Read more »

“Half my advertising is wasted, I just don’t know which half.” – Jon Wanamaker

It sounds like not much has changed.

To be honest, though I appreciate what retailers are hoping to accomplish, I think the situation is hopeless (one might even argue less diplomatically it’s naive). A potential customer is bombarded with multi-media campaigns — e-mail, banner ads, TV/radio/newsprint ads, etc. — at some point they give in and buy something. At which point did the surrender occur? On the last point of contact, or was it the first and they just waited…or was it somewhere in the middle? I don’t think anyone will ever know because I don’t think the customers themselves really know.

My instinct tells me that I would not design and execute a campaign that I could not measure. If the methods to measure are not direct, I would at least create a rationale that can be applied to create a “best guess.”

Unique codes and landing pages seem obvious ways to connect the channel with trigger of the promotion. The technology that enables the promotion should in most cases also provide the framework for measurement.

Is the problem related to poor retailer process or is there truly a technology related barrier to campaign measurement?

When I visit Harbor Freight, I give the cashier my member ID, and then I get the current promotional discounts. I receive emails, inserts, and see the occasional and magazine web ad, so which of the initiatives get credit for the sale? That’s just one example of campaign response attribution retailers face thousands of times a day.

My advice: Keep merchants and marketing closely aligned, understand and track select, meaningful KPIs, and watch for and tend to the profit-depleting exceptions.

I was just at a conference discussing attribution — this cross-channel measurement. And, I’m a mathematician by original training who has spent a couple of decades measuring ad campaign impact. My conclusion: We’ll never get the level of accuracy in cross-channel attribution that we need. There are plenty of vendors claiming to have magic black boxes that “reveal all.” But I’ve worked with those. They all contain huge statistical leaps of faith that the vendors are hesitant to reveal. And, as a result, are hugely unreliable. So what to do? Business is an art as well as a science. Our jobs are to gather the realistic evidence we can without putting ill founded faith in it. Then use our instincts to make decisions, watch what happens, and change based on what we see. (I will quibble with the article. There is significant evidence emerging that online is NOT hugely profitable for retailers. One article I read suggests there’s a 25% profit hit online when compared with stores.) And the idea that these customers are “more loyal”… Read more »

I’ve often found that this is the result of mismanaged consumer communications, lack of understanding that not all social media sites have the same function, and multiple agencies that don’t communicate with each other.

It’s not a complicated process. It’s possible to properly measure the appropriate tools. Also, it’s necessary to recognize that not all social media sites being utilized are appropriate or will deliver measurable results.

The best advice I can give any retailer is to consolidate advertising communications and charge those responsible with finding efficiencies in all media.