CRM used to mean “database.” Not anymore.
The New CRM unlocks the value of customer data. It leverages insights across every decision and marketing touchpoint and enables teams to optimize campaigns around lifetime value. This is CRM built for today's business-to-consumer marketers.
Seven of the top 20 US retailers use Custora to organize their customer data, build advanced segments, and deliver more personalized communications. More
Custora is a business-to-consumer CRM platform built to help marketers leverage customer data across marketing systems and tools.
First, Custora connects to all your customer data and uses predictive analytics to surface individual customer insights. Custora then provides marketers with direct access to build segments, answer complex customer-centric questions, and stream user-level insights across any marketing channel.
Over 100 leading retail brands, and 7 of the 20 largest US retailers, use Custora to acquire and retain valuable customers and improve overall customer lifetime value.
Over a hundred retail brands (including 7 of the top 20 largest US retailers) use Custora to acquire and retain high value customers and improve customer lifetime value. We power over 200 million targeted communications per month, analyze over 500 million customer records, and handle over $200 billion in transaction volume daily.
Crocs used Custora’s high value customer lookalike targeting on Facebook to drive a 10x improvement in ROAS.
Eloquii used Custora’s predicted churn segmentation to reduce customer churn by 27%.
Crocs used Custora’s promotion sensitivity models coupled with our web personalization integrations to improve profit margins by 2%.
The Dermstore marketing team used Custora to better understand customer behavior and drive an overall 2% increase in revenue growth. Whoa.
Tiffany & Co. used Custora’s predicted lifetime value segmentation to improve the performance of their direct mail catalog by 17%.
Calendars.com used Custora’s predicted affinity segmentation to improve ROAS on display advertising by 10x.
Lucky Brand used Custora’s predicted CLV for their direct mail catalog and generated an incremental $1.97M in revenue.
Tory Burch used Custora’s predicted fragrance affinity models and generated an incremental $825k in sales.
- Deep retail knowledge (and a product purpose-built for retail).
- A marketer-friendly interface with a natural language, sentence-based query builder.
- Guidance and prescriptive analytics (we tell you where to focus).
- Transparent + collaborative models and insights (we’ll happily explain what is in the box).
- Our customer success team are strategic consultants.
- Channel agnostic: We are a system of coordination and don’t replace your existing systems of execution.
Your analytics team is top notch and can generate rich customer insights, but they’re too overwhelmed with requests. Insights get stuck. We call this friction in the marketing department, or, “farketing.” When entire marketing teams can access customer data, generate insights and leverage them in campaigns, beautiful things happen.
ROI Calculator coming soon...
When discussing models, we typically open with some sort of Zoolander joke. Not this time.
There are a wide variety of retailing phenomena to model: When customers buy, what they buy, how much they spend, where they come from, and so on. Custora uses a variety of different statistical and machine-learning approaches for these different phenomena. Our core models lean on general optimization techniques (such as EM and stochastic gradient descent) over functions we have derived from models we’ve picked specially for their respective purposes. But in different areas we also use more conventional out-of-the-box techniques, like collaborative filtering or regression. If you’re ready to talk more sophisticated data science, let’s dance.
We’re interested in helping you identify what makes each of your customers unique. Accordingly, we emphasize making differentiated predictions on the level of the individual customer over targeting aggregates. These individual predictions can be sliced, summarized, and compared according to flexible criteria, based on observed data or even other predictions.
For example, you could look at your at-risk customers (predictive) by acquisition channel (observed), or examine the types of brands (observed) your high-value customers have bought (predictive).
Our modeling philosophy tends towards Bayesian approaches. As we observe each customer’s behavior over time, we revise our opinions about his or her probable future behavior.
We some some t-shirts that say Pareto No Big Deal. If you get this joke or even if you don’t and just want a free tee and have made it this far in the FAQ let us know and we’ll send you one.
Custora uses machine learning and predictive analytics to create a likelihood to churn score for each customer. The score is calculated and updated daily, so that shoppers that typically only purchase at holiday aren’t identified as at-risk if they haven’t made a purchase in 30 days, but weekly shoppers will be flagged after only a few weeks of non-activity. Retailers send tailored communications to shoppers when their churn risk score passes a pre-set threshold which maximizes the likelihood of keeping them in the fold.