AI, startup hacks, and engineering miracles from your friends at Faraday

Customer insight fundamentals: Building a framework for access and activation

Our customer insight fundamentals blog series aims to unpack important components of effective customer data analysis, prediction, and activation strategies. This article was curated from Robbie Adler, Co-founder and Chief Strategy Officer at Faraday.

customer-insights-activation-framework-blog

I feel like "turning insights into action” is the one of the most overused terms when it comes to marketing software data science services. Truth is, data and insights are relatively useless if your teams aren't prepared to act on them. So perhaps it's a bit ironic, but I want to focus on some critical learnings we've picked up around deploying customer insights and predictions at scale.

Set the right foundation and expectations

If you’re going to make an investment in machine learning and seek actionable insights from your customer data, you should not be looking for quick wins, done on the cheap. This doesn’t mean you need a team of data scientists or have to bite off a six or seven figure annual commitment, but it does mean you should approach the effort strategically.

At Faraday, we’ve spent a lot of time identifying what's needed at the foundation of an effective data science strategy for consumer brands (my colleague Tia covers data requirements in detail), but as a quick summary, we’d bucket them as:

  1. Identify your data sources.
  2. Identify the questions you are seeking to answer and why. Curiosity should not be the driver of these questions, but, rather, a clear path to action: "If I knew X, I would do Y."
  3. Identify the members of your team who will need to be involved in and supportive of the effort. In our work, these team members are most commonly involved in performance and field marketing, customer engagement, marketing strategy, and data science.

Build a solid data science framework

Once you have the right foundation in place, it’s critical you take the necessary steps to build on this. My colleague Bill chronicles how we approach predictive customer analytics for our clients, but to summarize:

  1. Survey your data. Avoid the “garbage in, garbage out” problem.
  2. Validate your models and insights prior to activation.
  3. Establish the integrations you need to support ongoing learnings and ease the path to activation. For us common integrations are data sources (e.g. data warehouses, customer data platforms (CDP), and CRM systems) and data destinations (e.g. ad platforms like Facebook and Google, marketing automation systems, and CDPs).
  4. Democratize access, so you’re not solely dependent on a unicorn, otherwise known as a “data scientist,” for activation. Good integrations inherently facilitate democratization. One small test of whether you’ve democratized access is whether access requires knowledge of SQL, R, or other programming languages. If so, you haven’t democratized access.

Audit team access and business impact

So, it’s finally time to activate. How can you tell if you’ve built the right structure?

  1. Are predictions and insights available in systems that your teams use in their day-to-day work? At Faraday, we’d want your performance marketer to see high-propensity audiences in Facebook, while we’d want your customer engagement marketer to see customer personas and product recommendations scores in their email marketing platform.
  2. Are your predictions and insights being clearly leveraged to guide strategy? At Faraday, this often manifests as our clients personalizing ad creative and copy, deploying promotions strategically vs. broadly, launching new products targeted at a specific subset of existing customers, and/or entering new markets based on predicted propensity of its inhabitants vs. qualitative bias or preference.

Learn more about customer insight discovery at Faraday and check out our integrations.

Customer insight fundamentals: Understand your customer data to make better predictions

Our customer insight fundamentals blog series aims to unpack important components of effective customer data analysis, prediction, and activation strategies. This article was curated from William Morris, Director of Data Science at Faraday.

Customer insight fundamentals customer data survey for predictions blog

At Faraday we make the distinction between descriptive statistics and predictive statistics. The latter frequently gets all the attention (“Seeing the future!”), but it can’t happen without the solid groundwork of the former. You have to understand what data you have at hand before you can leap into the predicted unknown.

The data survey

importance of data preparation quote

Customer insights — as practiced by Faraday — are an example of descriptive statistics, but there’s a lot more to consider when laying your data foundation. Most data scientists practice some version of what could be called a data survey; the goal is to surface meaningful patterns, gaps, and anomalies with an eye toward prediction. One of the more common ways for a data scientist to approach a survey is with a notebook, which allows for an exploration narrative, almost like a blog post with code and charts.

data survey notebook

For better or worse, notebooks can be freewheeling. While there’s no real limit to how far you can dig into certain datasets, at Faraday, there are a collection of standard metrics we look at in the process of building insights:

Acquisition time series

faraday recent orders persona screenshot

The pattern of how a group acquires members is usually essential in a data survey. This shows us if there’s a general upward trend, slowing enthusiasm, signals of seasonality, or spikes associated with specific actions like marketing campaigns or strategic discounting. A time series can also serve as the basis for any forecasting analysis.

Geographic distribution

united states data distribution map

Geography is a crucial indicator in a data survey. If a group is highly-concentrated in one region, it may not be the best seed from which to grow a national predictive model. The United States is a panoply of economic, racial, and cultural diversity, and geography is often the uniting factor in a host of demographic variables and indicators. Geography is also frequently the canary in the coal mine of statistical bias, and a good starting place from which to examine the implications of any future predictions.


Profile departure from baseline

customer insight comparison chart

As a component of Faraday’s customer insights analysis, we look at differences between a target group (e.g. “customers”) and a baseline group, like the whole - U.S. population. This allows us to gain a sense of the general profile of a customer and what makes them unique.

Next steps

Armed with a sense of what makes your data tick, you can confidently approach predictive analysis. Insights and surveys offer the full view you’ll need when adjusting a Bagged Decision Tree for regional weight, or tuning seasonality in a Prophet forecasting model.

Learn more about Faraday's customer insight discovery solution.

Customer insight fundamentals: Data requirements for Customer Insights Reports

Our customer insight fundamentals blog series aims to unpack important components of effective customer data analysis, prediction, and activation strategies. This article was curated from Tia Martin, Director of Customer Success at Faraday.

Customer insight fundamentals data requirements blog

From aligning creative and messaging based on who your customers are, to monitoring how your customer base is shifting as it grows, having a solid understanding of the individuals that make up your customer base is critical to maintaining successful growth strategies.

What is a Customer Insights Report?

At Faraday, we generate a wide range of customer insights and deliver them to our clients in the form of Customer Insights Reports. These insights are interpretations of trends in human behavior over time. They are intended to be both informative and actionable.

I’ve coordinated customer insight discovery projects for dozens of consumer brands — it all starts with getting the right data together. Before diving specific data requirements, let’s take a look at how Customer Insights Reports are created.

What can Customer Insights Reports tell you about your customers?

Customer Insights Reports can reveal a wide range of meaningful trends and patterns about your customers. Some analyses include identifying what makes your customers stand out from the greater population or specific geographies, patterns in product preferences and shopping behaviors amongst key cohorts, what differentiates one-time purchasers and loyalists, etc.

There are a variety of ways in which companies can approach developing a Customer Insights Report, beginning with a qualitative approach that would include surveys and direct interviews. Another way to approach this would be through quantitative analysis, using factors such as actual purchase history or financial information. These methods can be used independently or combined to help strengthen marketing strategies.

Faraday focuses on the quantitative approach, by combining first-party customer data, (think purchase history) with third-party consumer data, (demographics, purchase history outside of this company, and more) to develop a holistic picture of who these customers are.

These reports are customized based on what types of first-party data are available, as well as the kinds of insights that are meaningful to your current objectives, as well as your business as a whole.

Data requirements for Customer Insights Reports

Third-party data is necessary to expand the breadth and depth of your customer insights. We’ve built our own consumer identity graph which is comprised of nearly 300 million U.S. consumers and includes demographic, property & purchasing data from about a dozen sources.

When it comes to developing a Customer Insights Report, the first step is getting the right first-party data to match into the Faraday Identity Graph (FIG), which allows us to enrich the data you already have on your customers with hundreds of additional attributes.

In order to match first-party customer data into FIG there are a few basic fields that are required, these include:

  • First Name
  • Last Name
  • Physical Address
  • Phone (optional)
  • Email (optional)

Technically that is all we need to match into FIG. However, when it comes to generating meaningful insights and fully leveraging our prediction platform, more data is preferred.

Additional first-party data required for deeper analysis

Specific information about your customers such as when they became a customer and what they purchased will allow us to build predictions and provide insights on specific behaviors and actions future customers will take.

Below are some examples of the types of additional data that help expand the depth of your insights:

  • When someone became a customer
  • Items purchased
  • Amount spent
  • Number of purchases
  • Transaction history (purchase dates, order value, products purchased, etc)
  • Product(s) purchased
  • Discount used
  • Purchase amount

The rule of thumb is the more data the better. The more information our models can train off of, or the larger data set to glean insights from will allow for much more meaningful results versus just being based on a random or predetermined set of data.

Learn more about Faraday's customer insight discovery solution.



Customer retention requires data-driven predictions

Customer acquisition is a large focus for nearly every brand, and while this is important to driving growth (especially for early-stage companies), there comes a point when return on ad spend plateaus. To scale revenue and drive profitability, companies need to shift their focus to improving retention rates and increasing the average lifetime value of their customers.

After all, acquiring a new customer can cost five times more than retaining an existing customer. On top of that, approximately 80% of your future revenue will come from only 20% of your existing customers — so how can you find more of that 20% and further engage them?

Focus ad spend on high-LTV audiences

An effective retention strategy is inextricably linked to a strong acquisition strategy. If you can acquire customers from the get-go who are highly likely to come back repeatedly, you’re extending the reach of your marketing dollars far beyond that first purchase.

One way to identify leads and prospects who are likely to become high-LTV, loyal customers is to analyze what your existing loyal customers look like. Perhaps loyalty hinges on how often someone purchases or how much they spend; but if you dig a little deeper leveraging third-party data with your proprietary data sets, you might uncover even more telling details that serve as indicators of LTV.

From there, you can generate and target high-LTV lookalike audiences to increase the likelihood that those new customers are on track to become your next loyal customers.

Get ahead of churn-prone customers

Churn is a challenge for many brands that rely on repeat purchases to drive revenue. Luckily, the democratization of machine learning has made it easier to predict churn.

In fact, improving your retention rate by just 2% is the equivalent of cutting costs by 10%. This is because retaining existing customers is much more cost-effective than spending time and money on strategies to acquire new customers or to reactivate lost ones. So if you can accurately predict and then prevent churn by getting in front of risky customers before you lose their business, you'll be doing yourself a long-term financial favor.

When it comes to making accurate predictions about which of your customers is churn-prone, rich and informative data is key. Faraday uses machine learning models to find patterns and similarities between those customer sets in order to identify churn-prone customers. As your customer base grows, reiterating these analyses can prove helpful, as you might see shifts in behaviors and overall profiles as time goes on.

But knowing who is likely to churn is useless if you don't have a plan to act on that information. Those predictions need to be easily accessible in your ESP to segment your risky customers and put engaging promotions in front of them.


Churn modeling case study

Invest in personalized engagement tactics

Understanding and catering to your churn-prone customers is a financial investment, but investing the time and money into analyzing and personalizing your methods of engagement — and improving them by even 5% — can increase your company's profitability upwards of 25% and save you money in the process.

But it's not just the risky customers you should be concerned with when it comes to personalizing marketing creative. 89% of companies believe that customer experience is a key factor in driving loyalty and retention — giving your other prospects and buyers the same amount of attention helps to build trust and loyalty from the very start.

Personas are a great place to start when you're beginning to personalize your creative for various audience segments and customer groups. They can be high-level and just based on transactional history, or you can flesh them out with additional consumer data to create richer profiles to work from. Either way, providing more relevance in your communications with your customers shows your target audiences that you're putting effort into aligning with their interests, values, and motivations — ultimately building trust.

The loyalty that results from personalized customer experiences directly translates to the probability of a customer referring a friend to your brand. Neilson reports that 92% of consumers trust word-of-mouth referrals over all other types of advertisements, making loyalty programs and referral perks an incredibly valuable marketing opportunity (and a tactic many successful brands rely on).



Learn how Faraday helps brands maximize customer value by putting data science to work.



Four ways to extend your acquisition spend with data science

Acquiring customers who are enthusiastic about your brand and will support you in the long run is vital to establishing the scalability and longevity of your business. As the first step of building out a loyal customer base, customer acquisition is a high priority for marketing teams.

This article explores four ways you can leverage data science to extend your acquisition spend further to build a loyal customer base.

1) Build high-LTV lookalike audiences

The fact that it costs more to acquire new customers than it does to retain existing ones is no secret, which makes finding high-LTV prospects paramount to ensuring a successful, cost-effective acquisition strategy that will fuel future revenue.

Knowing your target audience is a key piece of acquiring valuable customers from the get-go. What makes your best existing customers unique? What draws them to your brand and/or products? Defining and identifying high-LTV customers is the first step in predicting who out there will become a valuable customer.

Predictive analytics play a crucial role in identifying traits and patterns in your customer data that are indicative of high-value customers. Of course, the depth and breadth of your customer data affects how accurate these predictions will be. Introducing third-party data is a great way to gain deeper insights, fuel your predictions, and generate higher-performing lookalike audiences.

2) Target lookalike audiences across multiple channels

Acquisition campaigns are most effective when orchestrated across multiple channels. Reaching likely buyers with consistent messaging through paid social, direct mail, and other targeted channels helps reinforce your brand in their minds and gives them more opportunities to engage.

To reach the same consumers across online and offline channels, you can't rely on anonymized audiences. Instead, you'll need a secure way of building known-identity audiences bolstered by third-party data. Faraday's predictive targeting tools let you rest easy knowing that the people seeing your ads are the leads you truly want to get in front of.

3) Optimize your channel mix with known-identity attribution

A diversified marketing mix allows you to reach varying demographics across different online and offline channels. But the sheer scale of marketing across numerous platforms doesn’t guarantee a high conversion rate or substantial ROI, so knowing where your money is best spent should be a high priority.

Equip your team with the capabilities to closely monitor campaign performance across each channel so you can identify what is working and what isn’t. It’s worth noting that this can be a difficult task with cross-channel campaigns, as several digital channels provide anonymized or aggregated results.

There are several attribution models out there, but we've found that known-identity attribution can provide you with a better understanding of who you should be targeting on each channel. Acquisition spend is precious, and having the ability to — and being excited about — shifting gears to improve your strategy is crucial. Marketing strategies should not be fixed; they should evolve as your customer base grows and changes.

4) Deliver smarter promotions and discount offers

Some consumers need a bigger push than others to convert — this is where promotions and discounts come in. But to protect ROI on your campaigns, it's important to predict which customers are will stay engaged after that initial purchase. Delivering a 15% discount to all consumer might help drive conversions and reduce your average cost-per-acquisition, but if they don't come back, you might find yourself in a tough spot down the road.

Similarly, not all consumers need a discount to nudge them along the path to conversion. Using predictive analytics to determine who should receive a discount and how big that discount should be can help you beat cost benchmarks and increase the overall return on your acquisition spend.

blog_cta_ogee-1