Mastering Attribution and Conversions

Technology has given marketers a diverse range of tools to develop strategies, create content, and reach out to their audience. With such a wealth of options, it can be easy for marketers to lose sight of which channels are performing well, and difficult to know which ones to prioritise in understanding and optimising the customer journey.

Marketing attribution - an organised set of rules assigning proportional credits for different touch points across the customer journey - has long been held as the gold standard for making these assessments. It’s a relatively simple (some would argue, simplistic) way of achieving oversight into marketing activities.

But as omni-channel marketing matures and we gain a deeper understanding of how various factors truly influence buying decisions, retailers may have to adjust their expectations and use of attribution modelling.

The Lure Of “Fake Attribution”

A 2016 study by AdNews and AdRoll found that 66% of Australian marketers were using some form of attribution. This figure jumped to 90% in 2016, according to a separate study by AdRoll. This second study revealed that first-click and last-click attribution models were the most popular, at 44% and 28%, respectively.

Though it remains a widespread practice, industry analysts have been talking about the flaws in attribution modelling for quite a while now. For example, touch-based attribution models work by attributing a sale to an ad that’s been delivered to a person before their purchase - whether they saw the ad or not. Many studies show that the majority of digital ads are never seen. To compound matters, the Association of National Advertisers has found that only one-quarter of all digital ad spend reaches actual people.

Melbourne Business School professor Nico Neumann speaks of “fake attribution” - flawed modelling that inflates digital advertising’s success when looking at sales, while overlooking more valuable metrics like long-term brand building (usually done by traditional platforms), simply because they’re harder to measure.

Common Attribution Mistakes And Misconceptions

There are several myths and misconceptions that lead retailers astray when it comes to marketing attribution. Here are the five most common:

1. Over-stressing the importance of the last touch -- or the first

As we remarked earlier, there’s been a traditional reliance by marketers on last touch attribution. But this model tends to give too much value to bottom of the funnel marketing activity, and too little to what’s occurring at the start and intermediate phases of the customer journey. A focus on the last touch point can blind marketers to the significance of other channels that are promoting higher conversions or sales. In many cases, this also deprives retailers of the potential to uncover previously untapped marketing channels and new sales opportunities.

Placing too much value on the customer’s first point of engagement with the brand poses the same risks.

2. Missing offline branding opportunities through targeting bias

In this sense, targeting bias involves focusing more heavily on the short-term activities that lead to a sale, at the expense of brand engagement at the top of the funnel. This kind of interaction often occurs offline. To overcome this bias, retailers should adopt a unified marketing attribution strategy capable of measuring offline to online sales.

3. Being misled by the quick movement of cheap inventory

Simple fact: cheaper goods are more affordable, so people are generally more likely to buy them. For this reason, low-priced retail goods tend to generate more engagement. However, this shouldn’t necessarily be taken as an indicator of the success or otherwise of your marketing endeavours, or the stickiness of your brand.

4. Measuring online and offline attribution separately

Considering your online and offline marketing activities as separate entities simply won’t fly in an omni-channel economy. Confining online and offline activities to separate silos for measurement results in limited visibility into all the interactions that occur across the broader customer journey. It also misses out on the nuances of what’s happening as on and offline engagements of the consumer affect each other.

5. Not taking cross-device attribution into account

Customers will engage with your brand using various devices, as well as different marketing channels. And with Instagram and other social media platforms taking a bigger role in selling, the range of platforms for shoppers to engage with retailers is expanding. Any marketing attribution strategy that doesn’t take cross-device engagement and interactions into account won’t be giving you the full picture.

A number of major brands have fallen victim to the misleading allure of attribution modelling for attribution’s sake - and they’ve had to adjust their practices to get back on the right track.

Rebel Aligns On Store Visit Conversions To Optimise Its Ad Spend

When Rebel, Australia's largest sports apparel and equipment retailer, wanted to discover the relationship between between search ad clicks and in-store sales, its efforts were hampered by siloed eCommerce and marketing teams, competing for resources.

To get a clearer view of the customer’s full path to purchase, Rebel measured store visit conversions in a combined on- and offline approach to driving overall sales. This gave a single view of the customer journey, enabling the brand to see the role that digital played in driving customers to its stores.

The organisation's omni-channel approach has paid off. Wayne Tozer, Rebel's executive general manager of retail says, "As we continue to drive a digital-first strategy for Rebel, we're seeing substantial sales growth on- and offline. Measuring store visit conversions has changed the way we look at our digital ad campaigns."

Total Tools Overhauls It Omni-Channel Marketing Strategy To Give 1100% Growth In Online Sales

Even with 81 stores nationwide and an online store featuring more than 20,000 products, professional tool retailer Total Tools was facing issues with a struggling marketing department, siloed information, and communications problems that prevented the brand from delivering the experience its customers expected.

A switch in the platform managing customer and digital engagement enabled Total Tools to totally overhaul its omni-channel marketing strategy. The new platform merges CDP (customer data platform), marketing automation tools, and CRM data in a central location, giving Total Tools a single customer view.

This single view enables the brand to personalise communications across multiple channels including in-store, email, Facebook, Google, and its own web platform. AI-powered insights allow Total Tools to deliver product recommendations and reach customers when they’re most likely to be responsive, increasing return on investment.

The result: an 1100% growth in online sales. In the two-and-a-half years since beginning the overhaul, open rates for emails (now the main channel of communication for Total Tools) have tripled, and click-through rates are up by a factor of eight. Social media has also climbed from less than 1% effectiveness to around 15%.

Adidas Readjusts The Balance From Overspending On Digital Advertising, To More Focus On Brand Building

At sportswear giant Adidas, a number of factors including reliance on last-click attribution, a lack of econometric modelling, and the failure to do any brand tracking swung the balance of marketing spend towards digital advertising. A focus on efficiency rather than effectiveness led the brand to over-emphasise ROI, and over-invest in performance and digital at the expense of brand building.

Long-term growth was also being inhibited by an over-supply problem, meaning Adidas products were too often sold on promotion, creating price sensitivity. Multiple agencies, inconsistent measurement, and a business set-up that had its main divisions competing against each other were creating friction on messaging and creative.

Over the past four years, Adidas has been working to change all this. With a new marketing playbook (dubbed ‘Creating the new’) and a renewed commitment to generating brand desire, Adidas has introduced a new campaign framework with emotional, brand-driving activity as its focus.

Adidas also adopted an econometric model that threw up some startling revelations. Whereas the brand had thought that loyal customers were driving sales (prompting investments in CRM), in fact 60% of revenue came from first-time buyers. Business units were not just driving their own sales - rather, all advertising drove general Adidas sales. And while Adidas had assumed that only performance drove eCommerce sales, it was brand activity driving 65% of sales across wholesale, retail, and eCommerce, while performance also drove wholesale and retail sales.

Those faulty metrics were caused by the brand’s reliance on four attribution models – Google Last Click, Google Custom, Adobe, and Facebook – as well as a focus on short-term, real-time measurements that focused on ROI and return on ad spend (ROAS).

Adidas is now working on creating the right mix of media and attribution models - a blend that will likely involve econometrics, as well as a test-and-learn approach.

Some Tips On Implementation

Actionable data collected and analysed in real or near real time is the key to creating the personalised and on-demand retail experiences that modern consumers expect. Generating and manipulating this rich data requires the monitoring and measurement of both online and offline channels.

In an omni-channel economy, understanding the role and value that each channel contributes to your acquisition and conversion events puts you in a position to optimise your marketing spend in the right channels. You’ll also be better positioned in planning future marketing campaigns, as the data available will assist in putting your focus on the channels that result in the highest potential return on investment.

With attribution modelling emerging as only one way of looking at a vast and complex ecosystem, retailers must be willing to cast a wider net in analysing the impact of their marketing activities. To do this, you’ll need to be prepared to suffer some short-term losses, and look beyond attribution to take a broader and more objective view of what’s really going on behind the scenes.

Marketing attribution and brand promotion will be hot topics at eTail Australia 2020, which takes place from 13-15 October, 2020 at Sydney Harbour Marriott, Australia.

15:40 CMO All-Star Panel: How can you implement a fail-proof and profitable attribution model that aligns with your business strategy to deliver profitable returns?

How does a marketing leader know when their organization is ready to adopt a multi-touch attribution model across their marketing org? What sorts of situations, problems, or goals reveal that it’s time to move forward with a more sophisticated attribution solution?

• How can you develop a model that properly weighs all the touchpoints that belong to each of your 5-10 channels to assign the correct amount of credit to the most important ones?
• Connecting marketing to revenue - How can you achieve down-funnel insights on which keywords, ebooks, articles, and emails that generate the most opportunities and customer conversions?
• “Single source of truth” - How can you leverage omni-channel tracking across all platforms to ensure data hygiene and conversions and leads aren’t double-counted?
• How can you align your sales team’s outbound activities next to your marketing strategy to accurately measure the impact of offline channels?
• How can you consolidate and integrate your offline and online touchpoints in a seamless fashion with your CRM platform to better inform decisions?
• How can you strategically craft your campaigns based on integrated, accurate, historic data that reveals the past and shows how to optimize for the future?

Some Tips On Implementation

---

Download the agenda today for more information and insights.