Freshplum pulls back its online promotions management kimono, reveals meaningful performance lift

By Michael Carney , written on October 28, 2013

From The News Desk

For online retailers, marketing is simply one big math problem. At the simplest level, cost of customer acquisition needs to be less than the lifetime value (LTV) of a customer. Not surprisingly, a lot of energy is dedicated to acquiring customers inexpensively. But there’s a natural floor in this category, while the upper bounds of the LTV figure is practically limitless.

Promotions analytics startup Freshplum helps online retailers effectively engage their customers through customized promotions. After nearly a year of offering its product commercially to prominent brands in the cosmetics, eyewear, electronics, apparel, footwear, home improvement, and department store categories – the company declined to name its customers – the company is revealing data on the effectiveness of its platform.

Freshplum is generating an average conversion rate lift of 7.32 percent, a 7.06 percent revenue lift, and a 6.4 percent margin lift, compared to control groups receiving non-customized promotions. The company has also been able to measure an average increase in repeat purchase average order value (AOV) of 38 percent. In short, it’s working.

The company was co-founded by former Facebook product manager and former  Divvyshot founder Sam Odio and former Cisco engineer Nick Alexander, with the aim of using data analysis to help its etailers more effectively personalize and target promotions to maximize effectiveness and therefore maximize revenue. Prior to Freshplum, etailers typically released blanket promotions to their members, such as $25 off purchase of $100 or more. But while this may prove effective for some customers, it misses the mark with others.

For example, some customers may only require $15 off to incentivize them to transact – in these cases, the etailer would over-discount. Other customers may require $30 off. And for some customers, a dollar value discount may be less effective than a percentage discount, like 10 percent off a single item of $75 or more, or simply a free shipping offer. Other common questions include those like, “Was our Memorial Day revenue attributable to the promotions we ran or due to an overall strong day for online commerce?” In each of these cases the use of a control group and Freshplum’s data science muscle.

“I paid my way through school with an ecommerce business, but I always struggled with pricing,” Odio says. “I could never find a good answer to whether a price was fair. When I spoke to a few friends struggling with the same problem, I knew it was something that there was an opportunity to solve.”

Today, Freshplum is all about promotion pricing, rather than item pricing, but this may not always be the case. To some extent, the company can glean insights about pricing through promotion effectiveness. For example, if a single item sells poorly, except when it’s heavily discounted, then it may be overpriced. And items that see limited lift from discounting, may be underpriced. But these second order distinctions are not Freshplum’s focus today.

Odio describes Freshplum as the outsourced data science team that the lower 99 percent of etailers cannot afford to have in house. Amazon, Walmart, Home Depot, Target, Macy’s, Staples, and a select few others have revealed publicly efforts to build in-house data science capabilities. But for the masses, it’s either use Freshplum (or a competitor), or rely on the blunt instrument of blanket promotions.

“Our best customer is a household brand that is highly effective with offline promotions but that is just learning online promotions,” Odio says.

Freshplum works by installing a Javascript API, similar to that of Google Analytics, that allows the company to observe how site visitors are reacting to an etailer’s existing promotions. The platform measures, AOV, buying frequency, lifetime purchasing volume, and other metrics to suggest new promotions strategies for each consumer or group of consumers. These promotions are then typically served on-site, rather than through email or other distribution channels.

Retailers often start with a single objective, such as converting “window shoppers” into buyers, or reactivating previously active customers who haven’t purchased in more than six months. Odio says that Freshplum may require as much as 90 days to collect data on user behavior and institute an effective campaign, the above data suggesting that the ultimate results are meaningful to a company’s bottom line.

“Most customers come to us because they want to be able to promote in some targeted way, like highlighting winter gear while it’s snowing outside,” Odio says. “It takes time for them to realize that the value proposition can be much broader than that. The relationships evolve and we typically end up becoming a more comprehensive solution for their promotions and customer marketing needs.”

While Odio wouldn’t reveal the names Freshplum’s clients, he did say that they combine to generate “several billion dollars in online revenue,” of which his company observes and manages more than $150 million worth. The company works with brands and online retailers directly, rather than with agencies.

Access to the promotions management platform is sold on a subscription basis, with pricing based on traffic volume. Clients get access to a basic Web console through which they can monitor promotions and also to a Freshplum account manager to assist in designing new promotions around the platform’s recommendations.

Freshplum has an eight person team today, comprised mostly of ex-Facebook and DemandTec engineers, Odio says. The company has raised a total of $1.8 million to date – only $1.4 million of which has it previously announced publicly – and has enough runway through the end of 2014, according to its founder. The company plans to raise a “large Series A” in early 2014, Odio says. Existing backers include Google Ventures, Charles River Ventures (CRV), New Enterprise Associates (NEA), Greylock Partners, Data Collective, Y Combinator, former Gmail product manager Gabor Cselle, Reddit CEO Yishan Wong, Greylock’s Alison Rosenthal, and CRV’s George Zachary.

There are other companies using data science to drive effectiveness in ecommerce, but none specifically focused on the on-site promotions category, according to Odio. For example, Santa Monica-based Retention Science offers customer retention and reactivation solutions through a data-driven email marketing platform. Elsewhere in LA, Steelhouse offer on-site promotions tied to retargeting and look-alike marketing campaigns, as well as other customer acquisition strategies.

As the ecommerce market grows increasingly saturated, there is mounting competition for limited consumer dollars and attention. Online marketers rely on a Swiss Army knife’s worth of tools to effectively acquire, engage, and retain these consumers. In many cases, however, these tools lack the deep data science underpinnings that they demand.

Longer term, the risk is that data science grows commoditized like SEO and SEM have over the last decade. But until there are more trained data scientists and companies beyond the WalMarts of the world can afford to hire them, there will be no shortage of demand for these services.

But today, that’s not the case. And 7.06 percent revenue lift and a 6.4 percent margin lift is real money that Freshplum’s customers can take to the bank.

[Image via MardenKane]