The missing link: online ads to in-store sales
Similar to how click tracking measures a web ad’s impact on online sales, a program being piloted by Google is designed to help retailers better figure out if their online ads are driving in-store sales.
According to a report in The Wall Street Journal, six advertisers are participating in the program with in-store data provided by Acxiom, DataLogix, LiveRamp and Epsilon. Michaels Stores was identified as one of the retail participants.
Through an AdWords feature called In-Store Attribution Transaction Reporting, purchases can be linked back to users’ cookies, which are matched to anonymous Google click IDs. Google can then tell which ads generated in-store sales, and how much they generated.
Currently, AdWords’ reports for advertisers include data such as click-through rates, cost per click and post-click information, such as when an online purchase then occurs. Under the pilot program, a column will point to a search’s impact on in-store purchases.
Google will reportedly pass an anonymous "click ID" through to the advertiser which typically has a cookie set in the shopper’s web browser so, in essence, they can identify who that user is and match the click to related purchases that occur later in-store.
Facebook and Twitter have also recently introduced projects with data providers to show the offline sales impact of their ads. But Google, as the dominant online advertiser, stands to significantly benefit if it can provide better metrics around in-store conversion.
Ginny Marvin wrote for Search Engine Land, "While Google has been introducing new ad formats and features such as image extensions to appeal to brand advertisers, the company knows that brands will really take notice if they can show that a search ad clicks or display ad impressions lead to in-store purchases."
Discussion Questions
If it proves to work well, do you expect Google’s “In-Store Attribution Transaction Reporting” to have an impact on how online advertisers do their work? Would successful click-to-store reporting shift more ad dollars to online?
I have always believed that you cannot manage anything unless you can measure it. This is a way to determine if money being spent provides results. So, yes this has value to retailers. It will be a good spend if it increases in store traffic and or sales. If not then spend the money elsewhere or not at all.
This is interesting for brick and mortar stores and ominchannel retailers, as it’s likely that folks who shop multiple channels spend more. This metric will give a “web-to-store” snapshot showing if online advertising drives in-store sales or not, which will not only help justify marketing spend, but also help understand shopper behavior.
Online to offline attribution is an important element in helping retailers optimize their marketing mix. At the moment, Google is attacking the problem in two ways. They provide “estimates” of offline influence and are trying to get advertisers to talk and think in terms of digital influenced sales vs. digitally attributed sales; and they are piloting true online to offline attribution.
The challenge with the current methodology is the data is limited. Only a subset of retailers share data with the data providers; only a subset of the transactions of those retailers use a traceable tender type (credit card); because the data providers get only part of the magstripe data, only a subset of the transactions are unique enough to identify; and only a subset of those unique transactions are linkable to a known cookie. With all those limiting factors, the Google test can only match to a very small percent of retail transactions.
But it’s easy to imagine that most of those problems go away in the future. As retailers move to more secure End-to-End encryption methods for credit cards, they can simultaneously generate tokens that are unique enough to match to individual households.
Retailers want to put their marketing dollars where they earn the highest returns. Given that only about 8% of retail sales happen online, it’s clear that understanding the offline influence of digital advertising is critical to optimizing the marketing mix.
For the first time, advertisers spent more in online ads than they spent on broadcast TV in 2013. As this trend surely will continue to accelerate, tools similar to Google’s and others available today need to be leveraged in order to determine ROI for such huge expenses.
The connection of advertising and in-store purchases has been a struggle for years. There are still a bunch of “soft” factors that enter into sales reporting, and you need to be able to filter these factors out in order to ensure accurate sales lift performance measurement. As these tools prove to be accurate, their impact on both retail and CPG brand marketers and advertisers will be enormous.
A missing link has always been how to track and measure the effectiveness online marketing with physical retail. What can be measured can be managed and improved. If Google can create metrics to measure online ads to store sales reliably, marketing dollars will certainly be more targeted to those online marketing channels that produce results.
However, the purchase journey of today’s Omni-channel consumers is neither simple nor easy to track. Consumers typically search at least 2.5 sources, as well as social media, before making a decision on a considered purchase. An interesting question arises as to which online ad or event was the “tipping point” that led to the sale.
If – if – it proves to work well, of course this will shift more ad dollars online. That being said, connecting online and offline in this manner:
“Google will reportedly pass an anonymous “click ID” through to the advertiser which typically has a cookie set in the shopper’s web browser so, in essence, they can identify who that user is and match the click to related purchases that occur later in-store.”
Should be challenging. It appears to be an exercise in mapping a browser’s identity to an in store purchase at some later date. If much of the work falls to the advertiser/retailer, this will take longer to mature, so hopefully Google is taking on more of this to help facilitate its adoption.
This is for sure a very significant part of the Convergence of Online, Mobile and Bricks (COMB) retailing. There is a necessary “web” that needs to transcend each of the boundaries, and Google seems to be in the catbird seat, although far from being the only player. I note the recent affiliation of Google with comScore, and also Catalina, in turn linking with Nielsen. Both comScore and Nielsen are of course primarily METRICS companies, comScore online and Nielsen in-store as well as TV et al.
I call attention to this because a lot of the early excitement about online advertising was driven by advertising companies (Google, Facebook, etc.) essentially vouching for their own ad products. That just is not going to work for any serious ad model going forward. The billions of dollars of ad money sloshing around looking for a place to be spent, needs solid metrics, independently validated. Hence the significance of comScore and Nielsen.
However, note that the Nielsen PRISM initiative crashed and burned, in my opinion, because those who controlled the media (the retailers themselves,) strangled any objective measurement of their “media,” if it wasn’t going to produce exposure numbers competitive with television. My scientific response to that is here: “The Incredibly Shrinking (In-Store) ‘Audience” and here: “From Opportunity to Final Purchase.“
Look for these same forces to play in the growing convergence mentioned above.
It sounds great in theory, but the devil is in the details. How do you give credit to an ad? Did someone click on it and if so, what was their intent? Does the ad really get credit for a sale if a customer only viewed the ad for two seconds? Customers have never said they wanted more ads to help them buy more stuff, they want better content.
This is smart on Google’s part and the other partners in the test. It’s also problematic that this is a problem yet to be solved in 2014. And it’s more evidence that even with digital advertising, like most paid media connected to the physical world (and even much of digital commerce), retailers – and their agencies handling paid media – still cannot adequately measure return on marketing investment when the “investment” is paid advertising.
It would certainly change the mix of advertising if online advertising can be attributed to in-store sales, which is currently done by estimation. Still need to understand how the mechanics work on matching the in-store purchase transaction ID back to the online ad, since the click thru isn’t as obvious on the last mile. If I click on an product ad and buy later, it is assumed that the ad drove the sale and not the flyer or freeway billboard I saw in between? Probably will need some fine tuning, but it is promising.
Connecting the dots – or maybe the clicks – is the next big step forward in retail measurement.
So far, this has been possible on small-scale campaigns, with participating retailers. However, access to connected data (online to offline sales) on a mass level has proved just beyond the grasp of most national brands.
Note – retailers, however, are sitting on more data – their website traffic connected to their shopper cards. How effectively the retailers have been using or sharing this data is another matter.
So – Google is taking a stab at this. It makes sense. It’s only a question of time before we have a 360 of shoppers’ exposures and behaviors, and Google wants to stay out in front. The future is algorithms, and Google knows it.
Absolutely, click-to-store would have an impact on online spending, since retailers would be able to track an increase number of visitors to actual transactions. In addition, retailers can execute real-time emails and retargeting based on behavior of a customer who is likely to purchase anywhere — store or online. The increased revenue attribution will impact online ad spending tremendously we know, since we know that high consideration purchases are researched online — we just have struggled to connect that research to the purchases.
That will be a tremendous breakthrough.
This is huge, for both advertiser and retailer. To actually be able to tell what works is a marketer’s dream. I would expect that if the results are definitive, good or bad, more ad money would pour into agencies that parlay this m.o., including Google.
My wife is an avid “see mobile display = run to store” shopper as I would imagine are many others. To quantify that, again, is golden.
Side note: love the fact that Google is looking to solve age-old retail problems vs just selling stuff. Solid, long-range thinking.
The design of a usable online/digital coupon that describes place or means of origin with built-in lifespan or void is a better fit for this need. Software developers could make this happen, but a far more complicated solution often pays better in both front end sunk costs and residual fees.
This is how retailers that have the foresight and means to create an in house application development plan will get ahead of the current trend of buying more software with less payback. Allowing the company-owned development group to copyright and license the software to other retailers could assist with managing costs. The limited range of awareness in Information Technologies held by a large majority of CIOs is largely at fault for the resistance to expand in this or a similar direction. We may still be one to two leadership generations away from endeavors like this.
Sounds interesting but some missing dots for me – how will attribution be modeled? If I click on an ad online and subsequently go to the store (tomorrow, next week, next month, next year, etc.) and make the purchase – how will the ad get credit? Wouldn’t I need to be identified at the till somehow?
Moreover, I would think there should be some consideration of timing – if I clicked on something more than a month ago I don’t feel that the ad should get credit. Did the ad truly change (or even influence) my behavior, or did I click on the ad because I was already interested?
Seems like a huge gulf to be spanned between theory and practice here, but there is no doubt a huge opportunity in omni-channel marketing.
If in-store attribution transaction reporting proves to be as big a success as Google is playing it up to be, then yes, of course advertisers will shift a higher dollar spend online. This was bound to happen with the increasing number of consumers shifting their product research online (mobile, social, web, etc.). It’s great to hear that Google is taking initiative to bring the same level of measurement and quantification to in-store attribution as they did to online advertising.
Understanding what customers do in the store is vital to all brick-and-mortar retailers. This service will help close the loop between the customer’s activity inside and outside the store.