Optimising marketing channels at scale: Why and how?

We have previously discussed how marketing channel optimisation is contingent on two processes; data integration and attribution. The first article, on data integration, covered the importance of tracking, stitching and integrating customer data from the entire end-to-end customer journey in order to produce a granular single customer view. The second – multi-touch attribution – examined …

by | October 18, 2017 | Fospha

We have previously discussed how marketing channel optimisation is contingent on two processes; data integration and attribution.

The first article, on data integration, covered the importance of tracking, stitching and integrating customer data from the entire end-to-end customer journey in order to produce a granular single customer view. The second – multi-touch attribution – examined the subsequent need for financial services marketers to use this single view to accurately attribute the value of their cross-channel marketing activities, right down to individual keyword level. With both of these processes in place, financial services businesses stand much better equipped to reduce their cost of customer acquisition – which stands at a pricey $175 and $303 in the financial services and banking/insurance industries respectively (Kurt Kummerer, 2016).

However, the final step to achieving full marketing channel optimisation is then operationalising this data.

Why optimise at scale

Operationalising refers to marketers being able to take their data insights, and act upon them at scale. For instance, it is simple to move budget between two or three keywords once a week, but doing this every day and with multiple keywords is a large feat, and highly time consuming for one marketer to manage. Therefore, operationalising these insights in order to act at scale and in real-time is as much of a necessity as integrating data and carrying out the multi-touch attribution that provides these insights in the first place.

The benefits of operationalising data are unparalleled – for instance, at the Adobe Digital Summit in March 2016, RBS reported that their expansion of optimisation efforts has generated more than $1.5 million in incremental revenue. Clearly, if you want to optimise your marketing channels to increase revenue, it has to be done at scale and in real-time.

We’ve identified three main marketing channels where operationalising at scale can bring real value to a business: Paid Search, Paid Social and Email.

Paid search

The most effective way to operationalise paid search insights is through a bid management platform – such as Kenshoo or Marin. For those unfamiliar, these solutions manage, automate and optimise search engine marketing campaigns at scale, by automatically reallocating spend across all channels based on business goals. Marketers are often able to manage their budget to improve ROI, conversions or increase spend for a set outcome. However, as identified in previous articles, the key to optimising marketing channels is in harnessing your data and using multi-touch attribution, so that marketers can understand the precise value of every marketing activity.  

The answer to large scale paid search optimisation is to integrate a bid management platform with a Customer Data Platform. Although useful on their own, bid management platforms still run off a businesses’ data silos – meaning they only provide a limited view into your marketing channel spend, and one that is not always accurate as it fails to take into account the full end-to-end customer journey.

By plugging multi-touch attribution data into a bid management platform, marketers can ensure their insights and actions are fuelled by the cleanest and most accurate data, whilst taking advantage of the management and automation that the bid management platform provides. Under this model, Paid search channels are optimised in the same way as if using a bid management platform on its own– but the improved data source means that more accurate decisions can be made to drive more efficient results

Paid social

A second popular channel, and one that financial services businesses are increasingly turning to, is Paid Social. Facebook is a great place for advertisers to build trust, increase awareness, and turn first-time customers into lifetime advocates – with 61% of financial professionals with a Facebook profile already using it for business at least once a week (Blue Foundation). Despite the financial services industry being highly regulated and scrutinised, these strict guidelines don’t have to stop marketers from reaching their customers digitally. Facebook targeting enables marketers to find the right customer segment based on specific demographics like age, location and gender. Businesses can also target customers using their own data, in order to target Lookalike Audiences. However, as with paid search channels, Facebook’s inherent ability to increase conversions and ROI is hindered by its data source.

The reason for this is two-fold. Firstly, without a rich data source and the ability to attribute accurate value to paid social channels, marketers simply do not have the necessary visibility on what actions are leading to conversions. And, without this, it is hard to justify spend on this channel.

Second, paid social marketing is contingent on targeting the right customers with the right message. However, only so much can be achieved by targeting generic clusters with a generic message; the chances are that those you are targeted are not even interested in what you are pushing. The result is wasted paid social spend, and difficulty being granted a paid social budget when you are unable to demonstrate a positive ROI from this channel.

The integrated data in a Customer Data Platform – and the granular detail it provides of who customers are, where they have been and how they have interacted with your business – provides marketers with a wealth of information about their customers. This makes targeting much more accurate, as marketers can group potential customers on the basis of much more detailed information – such as having previously browsed for a mortgage but not filled in an application, or interacting with a banking app in the last 24 hours.

Much like paid search optimisation, the key is to take this data source and plug it directly the platform – in this case, Facebook Ad Editor. With the richer data that a Customer Data Platform provides fed directly into the social media platform, marketers are able cluster potential leads on the basis of highly specific behaviours, and automate relevant, conversion inducing content. This integration means marketers do not have to manually segment customers, but instead can use Facebook’s capabilities and their rich data source to automate targeting at scale and in-real time; based on the every changing single customer view the Customer Data Platform provides.


Much like social media channels, one of the main challenges in email marketing is the data source that automations are based on. With purely demographic information being fed in, clustering and personalisation is highly limited and ineffective. However, with the granular single customer view being plugged into email automation tools – such as Marketo or MailChimp – financial services marketers can harness machine learning to create customer clusters based on highly specific customer engagements.

The artificial intelligence layer found in a Customer Data Platform can then match relevant content to these segments, and push the email out in masse. The difference between this and manually sending single emails to high prospect customers is evident.

For instance, emails sent one hour after cart abandonment have been shown to boost conversions by 100% (Grin). Without a rich data feed that uses machine learning to identify customers performing unique actions (such as abandoning a cart) and to send an email based on certain parameters (such as after a one hour time frame) how are marketers meant to catch these specific circumstances that can boost conversion rates and ROI?


The message is simple – whilst rich data and a clear understanding of your customers and customer journey is priceless, being able to operationalise your insights at scale and in real-time is the next step in achieving seamless marketing channel optimisation. Marketers therefore need to focus on taking their rich, attribution data and plugging this in to marketing software to supercharge their automation and optimisation efforts. Without this, they are left with time consuming, manual work that will not bring about the marketing optimisation needed to boost conversions and ROI. Although this problem is not unique to the financial services, it is an area that marketers in this industry need to tackle in order to ensure they are not left behind in the digital marketing surge. 



Regulatory reporting: 7 Questions with Philip Flood, Gresham Technologies

Other | Behavior detection & predictive analytics Regulatory reporting: 7 Questions with Philip Flood, Gresham Technologies

Gresham Technologies
Real-time payments tech put pressure on banks

Best Practice | Behavior detection & predictive analytics Real-time payments tech put pressure on banks

Managed Services in 2021: Poised for Lift-Off

Best Practice | Behavior detection & predictive analytics Managed Services in 2021: Poised for Lift-Off

SmartStream Technologies