The Challenge with Social Media Advertising

You can’t drive a return on investment from social media.

It was a criticism that was widely levied at social media as a marketing channel not too long ago. Social was seen by some as far too passive a medium; a platform where ads weren’t effective, simply because the users of social media were hostile to them. And even if they weren’t, people would argue that those audiences would move around so quickly that it was difficult to target the ones that you wanted.

Additionally, limited targeting options and a lack of functionality added weight to that argument, but things have somewhat changed.

The business case for social ads

Facebook first launched its advertising platform in 2005 and, ten years later, social advertising revenue across the leading platforms for the US stands at $8.4bn. By 2018, it’s expected to top $14bn. Those figures are fuelled by a growing number of marketers investing in social as a channel. A survey of 5,000 marketers by Salesforce found that 70% were planning on increasing their social ad budget for 2015, whilst two thirds of marketers believed that social media was core to their business.

These sorts of statistics belie the suggestion that social media does not drive a return on investment. It may be an advertising medium that is somewhat in its infancy, but the statistics suggest that there are very real opportunities within social advertising.

The rewards for developing a social advertising strategy are becoming increasingly evident. The typical click-through rate (CTR) for a Facebook desktop ad is more than eight times higher than a normal web ad, and on mobile the CTR is more than nine times higher. (Source: Hootsuite)

On Twitter, promoted tweets average engagement rates in the region of between one and three percent – higher than traditional banner advertising. (Hootsuite). But social media advertising is a very different form of advertising to other forms of paid digital media.

Social media advertising cannot be seen merely as an extention of your existing paid strategy, and you cannot simply clone your strategy for one social platform and apply it to another.

Critical to the success of your campaign is your understanding not only of the audiences that use each platform and the subsets that you wish to target, but also of the platforms themselves. Each social network has its own technical limitations and, perhaps more importantly, its own distinct culture. What is acceptable to an audience on one platform may not be acceptable to the same audience on another.

But if you get that strategy right, social media can be a real, tangible driver of leads and sales.

Getting to know your audience

Facebook and, to a lesser extent, Twitter have extended the boundaries of audience targeting, providing advertisers with much more control of their ads, more granular targeting options and better opportunities to get in front of their core audience.

Users are the primary asset for social networks, and it is in the interests of each to protect the userbase from receiving branded content that isn’t relevant to their interests. This has prompted the social networks to create extremely detailed layers of segmentation to ensure that advertisers only target the most relevant users. This protects users from irrelevant ads and allows advertisers to reduce their ad costs.

What do your customers already tell you?

Your current customers provide unrivalled levels of data into how people engage with your brand, so using this insight should be the first step before developing any form of paid social advertising.

First party CRM data will provide a considerable amount of data regarding your current customer base, how active they are in particular product areas and how they engage with your brand. This data will help you target:

• Those customers that are actively engaged with your brand.
• New audiences.
• Audiences that may have previously transacted with your brand, but have failed to complete a purchase since.

The way in which you segment this data is ultimately what will determine how valuable it is in steering your initial targeting campaign.

Segmenting existing audiences

Segmenting your eCRM data correctly will ensure that you can drive an effective campaign. By focusing on target groups based on their profitability to your business, or frequency of purchase, you avoid investing ad budget in targeting people who have a low propensity to convert through to your website. Typically, this data would segment into five key areas.

Lifetime Value

This represents the total value of a consumer group over the course of the consumer lifestyle. The more valuable the customer group, the more you should target them.

Highest Basket Value

Highest basket value (and similarly, average basket size) identifies consumer groups that may not necessarily have the largest lifetime value, but who typically purchase more products, or more valuable products, in one purchase.

Most Frequent Purchasers

These are your most loyal or most active consumers, and are likely to be more receptive to your brand advertising campaigns.

Location Concentration

Does your brand have a audience that is focused around certain geographic areas? This can apply to both regional businesses and large multinationals, who may find strength in certain geographic regions, be that a particular group of cities, counties, states or countries.

Channel Acquired

Understanding where your customers have come from is equally as crucial. A user who has converted for your brand via online channels such as PPC, display or remarketing are likely to spend more average time online than phone or direct mail consumers. They will likely be easier to build an online profile for whilst also being more responsive to online social advertisements.

Targeting your audience

It would be easy at this point to dismiss this process as one that doesn’t drive new customers to your brand and, in many respects, that is true. However, what this process allows you to do is take first party data from your current segmented customer list, and run these through online audience profiling tools in order to understand fundamentally how your customers behave. It stands to reason that, unless your organisation is planning a significant step-change in the audiences that it targets, the audiences that you want to attract are likely to have very similar demographic and behaviour patterns to the ones that are already engaged with your brand.

Uploading this first party data, such as segmented e-mails or phone numbers, into social advertising tools allows you to instantly create new “custom audiences” within the platforms. You can then analyse these custom audiences to gain an understanding of how your customers from the list behave on social media platforms, and this will dictate the strategy you adopt. Typically, this data will be categorised into the following:

Demographic

This data will segment factors such as age, gender, relationship status and parenthood status.

Location

Provides a geographic breakdown of your audience. Depending on the scale of your organisation and the nature of your campaign, you could break this down by postcode.

Behavioural

This breaks audiences down by their online behaviour. This could include the device types they use and the times of day that they use them.

Economic

This highlights variables such as household income, profession, job title, social status and homeownership status.

Social

What groups or affiliations do these audiences have? Do they have higher or lower levels of mobility (ie, willingness and ability to travel)?

Interests

This category identifies your audiences interests, hobbies and activities. This is particularly useful for brands that target a particular niche.

This data can now be cross-referenced across each of the individual segments identified in order to understand where there may be areas of crossover from one consumer segment to another. If, for example, you find a correlation between income groups and the time of day in which they use social media, you can use this information to specifically target that income group at a particular time or on a particular device.

Building these personas of your existing audiences allows you to identify how similar audiences are likely to behave online. It is likely that the audiences you want to attract will behave in a similar way to the audiences that you already engage with. Deploying a lookalike campaign to target new audiences based on the behavioural and demographic traits of your existing customer groups means that you are specifically targeting audiences that are likely to engage with your brand.

By focusing on target groups based on their profitability to your business, you avoid investing ad budget in targeting people who have a low propensity to convert

You’ve found your audience, now it’s time to optimise

Optimisation of social media advertising is a fine art and at the heart of all optimisation activity should be your business objectives and KPI’s. The optimisation process is structured around these core metrics and through the granular segmentation of the account.

Looking at how the account is structured, and analysing this against your historical first party data in regards to performance, helps you to identify the correct bid method for each of the individual segments. Clustering each audience based on tangible behavioural activity is crucial, as this allows you to segment your audience to extremely granular levels. This gives you greater control in deploying campaigns that deliver a message to only those customers that will deliver an ROI. Analysis of each bid method and determining what the likely outcome would be from the deployment of that bid method ensures that your ad spend is being deployed in the correct way and in the correct places.

Structuring your activity

Stucturing a social advertising account is, in many respects, very similar to the process of structuring a paid search advertising campaign within Google Adwords. You should focus on grouping logical targeting at campaign level, such as eCRM targeting and conversion lookalikes, and overlaying these with specific products or categories based on the insight that you have for that particular target demographic.

The objective of the campaign must also be taken into account across all activity, as once this is set up this cannot be changed. Therefore, ensure all campaigns are grouped logically, with the objective of the campaign clearly stated (whether that is clicks, engagement, conversions, etc). In order to get the best out of this structure, we would recommend separating your ad sets based on age clusters, gender and device. In the example below, we have sampled the target audience for a branded trainers product line.

This provides advertisers a greater amount of granularity against which activity can be optimised. Additionally, this structure allows ad creative to be tweaked and made more relevant to each audience within an ad set. This will increase both CTR percentages and relevancy scores within Facebook, which helps in achieving the highest possible reach and helps to reduce delivery fatigue, where your audiences become tired of seeing the same ads repeatedly.

Remarketing

There are two core forms of remarketing that Stickyeyes uses across social media advertising campaigns for direct response activity. Both of these approaches are intrinsically linked and underpinned by data and granular segmentation, but target two very separate customer segments.

Remarketing to dormant consumers

Dormant customers can make up a a very important part of your consumer database. These are people who have previously engaged with your brand and have previously purchased but, for whatever reason, have not made subsequent or recent purchases.

Whilst it may sound counter-intuitive to market to these audiences, this data actually provides you with a wealth of information that can be useful in not only re-engaging those audiences, but also support your consumer demographic profiling.

Usually this process will look at your eCRM data and identify the last purchases a consumer made, combining this with other information such as the types of items / services purchased, and the previous frequency of purchase. This information provides a snapshot of your customers, which you can then begin to segment on a range of factors.

One such factor could be dormancy, namely the length of time since their last purchase. The less amount of time that a user has been dormant, the less incentivisation they are likely to need and the fewer touchpoints they are likely to use, so you should be spending less in attracting this audience. However, those audiences that have not engaged with your brand are likely to need more incentivising and more persuading. This could even include targeted incentives such as discount codes or exclusive offers.

Site visitor remarketing

Prior to deployment of any activity on social media, it is advisable to deploy custom audience tags (re-marketing tags) across every page of your website. These tags enable advertisers to understand how users behave on site, what the conversion journey is and where the ‘leak’ points of a website are (for example, does your delivery page create an increase in dropped baskets?).

Once we understand how users are behaving on-site, we can determine which areas of the site would be most effective at converting the traffic from your remarketing activity. This can typically differ based on vertical, or based on the complexity of your online proposition, but the below diagram highlights a typical remarketing list that would drive greater efficiency from activity on social media channels.

[Table or chart in PDF file - Register or sign in to view]

Taking into account the previous level of engagement of a user is important for three key reasons; areas, relevancy and optimisation. Ensuring that you are as relevant as possible to each of the customer segments through accurate and engaging ad content, combined with the most appropriate landing page, can help accelerate the user conversion journey and reduce the number of “potential” touch points from other online channels. This drives last click conversions through remarketing and lowers the lifetime cost per acquisition (CPA) required to acquire or re-engage that customer.

Taking the example of a consumer who views a product page, an advertiser would typically look to deploy dynamic based remarketing on social activity to remarket back to them with specific products they have either viewed online, or similar products within that range. Given that you know these customers have not made a purchase previously, or added a product to the cart, you could incentivise this with a “free delivery” on your first order message, or 10% off for new customers.

For a previous customer however, we would still recommend using dynamic remarketing based on the products that the customer has bought previously. Ensure from the feed that the product they purchased is not displayed to them (unless buying cycle is frequent) and focus on similar products, upsell products or products that “complete the look”. Given that they would not qualify for a new customer discount, incentivisation and messaging changes to be more focussed on “chosen for you” or “tailored to you” with an emphasis on pushing to customers first other exclusive offers or discounted products before opening these up to non-converting website visitors.

Choosing your ad format

There is no ‘right or wrong’ ad format to use, and ultimately different ad formats work for different brands, different audiences and different purposes. The table below represents a selection of ad formats, and this is something that you need to test and appraise in order to find the formats that work most effectively for your campaign.

Stickyeyes has found that carousel adverts have worked particularly well across our current client campaigns where we have utilised remarketing activity. This particular ad format allows advertisers to deploy up to five links on a rotating carousel, displaying five different products that they can click on as part of your advertisements. Crucially, this has the potential to drive them to five different landing pages.

However, this advertisement format is likely to be less effective at targeting new customer who have either never engaged with the site, or who are unaware with the brand. For these audiences, a more traditional single image format, or new video format advert may be better in order to immerse prospects into your brand and drive them through to site.

The case in point

Stickyeyes has implemented this level of customer targeting through Facebook advertising for a number of clients, including a fixed price fashion retailer that was looking to increase revenue and enhance customer growth through paid media.

As a fixed price retailer in the crowded fashion sector, the client was particularly susceptible to increased ad spend costs, so we needed to find a more cost effective biddable media route than the increasingly competitive and expensive AdWords.

Facebook advertising provided us with a clear opportunity to drive sales at a lower cost than other forms of biddable media. However, our success was dependent on our ability to interpret data and deploy ads to only those audiences that were likely to generate a return on investment. Our team therefore analysed several key data sources, including the client’s eCRM data, social media analytics and third party demographic data to deploy a Facebook advertising campaign that targeted the client’s core target audience. This analysis presented two key opportunities.

By targeting dormant account holders, who were already in the client’s eCRM as having previously engaged with the brand, we would reconnect our client to those audiences.

Additionally, by defining a clear customer profile for the client, we would use email look-a-like targeting, interest targeting and page look-a-like targeting to ensure that the client’s investment was directed only the most engaged consumer demographics. Overall activity consistently delivered a healthy ROI of 450% for the client, with 88% of sales volume generated from profile look-a-like activity.

Look-a-like targeting provides the lowest CPA across all the activity with an average CPA of £1.43 (compared with a previous average CPA of £10.18)), against an average order value in excess of £26.