For all the technological advances of our time, business software is still profoundly dumb. Most enterprise software simply serves as a tool to get a particular job done faster or more efficiently, but few applications are really “smart”. True smart enterprise apps leverage ‘Big Data’ and predictive analytics to spot patterns, draw conclusions and deliver insights. Most important, big data software learns and adapts itself to make end-users more productive. More of these smart big data apps are coming, however — and it’s about time.
I recently wrote a post called The Software Entrepreneur’s Playbook that detailed a range of strategies necessary to build a billion-dollar enterprise software company. One of the most critical ‘plays’ a software company must make today is to leverage big data and predictive analytics into its products.
Two key changes have happened in the market recently that have made this possible.
First, the amount of behavioral and transactional data available on consumers, companies, and trends has exploded over the past five years, as all of our activities are tracked on the Web and through our mobile applications. This has provided smart software entrepreneurs with a treasure trove of information that can help inform product development, sales, marketing and more. By leveraging this data, entrepreneurs can figure out which product features and benefits will be the most impactful to end-users — even before they’ve launched V1.0.
Second, the infrastructure to ingest, manage, query and understand all of that big data has improved dramatically. The innovation and cost compression occurring in cloud computing, storage, databases, data warehousing and business intelligence now enables entrepreneurs to collect and crunch data in a way that was impossible a few years ago — unless you were able to spend millions of dollars on your technology infrastructure, which of course cash-strapped startups could not.
The intersection of those two trends — a flood of big data and the infrastructure to make sense of it — is today providing smart entrepreneurs with a unique opportunity to invent new software product categories, or to re-invent and dramatically improve traditional software sectors. Here are three areas that already have been impacted.
Not long ago, say 20 years back, sales people mostly used spreadsheets, the phone and even faxes to get their jobs done — tracking leads, following up on prospects and booking meetings. Then, along came Salesforce.com to make the entire sales process easier and more streamlined. With Salesforce, sales reps could now enter and track contacts in one central database, improving coordination between teams when pitching or closing deals. But while Salesforce was certainly an improvement over spreadsheets and faxes, it is essentially a giant contact database — really a file cabinet — that requires a lot of manual data entry. It provides almost zero intelligence into the sales process to help reps close more deals and better serve customers.
By embracing big data and predictive analytics, Customer Relationship Management (CRM) software can provide a lot more than a static contact database and deal funnel. An intelligent CRM system would use behavioral and transactional data to tell a sales rep where to find new leads; which prospects have the highest chance of closing and when; which products each prospect is most likely to buy; how much they will spend; who is most likely to upgrade; which leads need an ‘extra push’ before they will buy; and other predictive scenarios. In other words, a smart CRM application would help sales reps do their jobs more effectively instead of burdening them with time-consuming data entry.
Thankfully, many startups are re-inventing CRM around big data and predictive analytics including companies like RelateIQ*, 6Sense*, Yesware*, Lattice Engines and Infer. Be on the lookout for these companies as they can help your sales reps close more deals and generate more revenue.
Human-resources software is also undergoing a big-data revolution. HR platforms like Taleo and SuccessFactors are much like Salesforce.com — static databases requiring HR managers to manually enter resumes and contact data. They essentially serve as resume repositories, but require HR managers to do all the heavy lifting themselves.
A big data-enabled HR platform is far more than a resume-management system. These new products, from companies such as Entelo*, Knack, and Evolv, can actually find the best candidates and predict their likelihood of accepting and keeping a job. By mining data in resumes, and comparing it to third-party data from platforms such as Facebook, Twitter, LinkedIn, Github and Stack Overflow, these systems can predict which candidates are optimal hires. Some big-data HR products take this a step further and use predictive analytics to scan the Web and find “passive” candidates who aren’t necessarily looking for work, but who would be great candidates for a company’s open jobs. Big-data analytics can show which candidate is the least likely to job-hop; which are the most dedicated and well-liked by bosses and employees; and which would be the best fit culturally for a given company or role.
Big data can also help HR teams better manage existing employees. Using this type of predictive software, an HR manager could find out which employees are most likely to quit in the next few months — based on analysis of first-party data such as an employee’s age and role, how long he/she has been at the company and recent performance reviews, as well as third-party data such as how many times an employee has checked LinkedIn or looked at job search sites.
For example, mobile security company Lookout recently needed to hire a security product manager, a very tough role to fill in the competitive San Francisco Bay Area. The company used big-data, HR software to find the exact right candidate. The software’s algorithm used predictive analytics and pro-active Web crawls to find potential candidates — even those not actively applying for jobs. It analyzed more than 70 variables potentially indicative of upcoming career changes, from layoff announcements and M&A activity, to potential candidates’ length of time at current company and social profile activity.
Sometimes, big data analytics uncover highly surprising findings. Evolv, a big-data analytics platform for workforce optimization, found that former criminals make great employees. Evolv calculates that employees with criminal backgrounds are 1 to 1.5 percent more productive on the job than people without criminal records. So next time, don’t be so quick to throw out that resume from San Quentin.
How can you make existing customers happier and less likely to churn? By deploying big-data enabled, customer-success software, of course. In the past, software companies sold multimillion-dollar perpetual software licenses, each carrying an annual 18%+ tax for maintenance. There was no notion of churn because the software vendor received all of the revenue up front, and it didn’t really matter to them if the customer was successful.
Now, however, cloud and SaaS delivery models have upended the entire software industry but have also introduced a new challenge. When customers can simply “turn on” a subscription to a low-cost, cloud-based application, they can just as easily turn it off and change to a competitor’s service. Thus, the ominous threat of churn is always hanging over today’s software companies.
Today, smart customer-success management software is helping many subscription-based companies dramatically reduce churn. These platforms can predict which customers are most at risk of becoming disgruntled; which are the most unhappy and why; which are just about to jump ship and go to another vendor; and which are the most contented and why. You can also use the technology to take a deep dive into stats for each customer based on account manager, sector, contract size, etc. Customer success software companies leveraging big data to transform customer relationships include Gainsight*, Bluenose, Totango, and Scout Analytics, which was recently acquired by ServiceSource.
A recent success story is a SaaS company in the healthcare industry that had 1,000 customers, but many were small businesses, so retaining them could be a challenge. The founders realized that churn could easily reach into the double digits, unless they used big data to find out how to combat customer loss. Using an intelligent customer-success management application, the company instantly saw which customers were on the brink of churning, why, and what could be done to keep them. In the end, the company reduced churn by 50% and generated 10% more leads through cross-sell, which ultimately resulted in over $2.5M in new annual revenues. And it did this with a team of seven, instead of the 30 people it would normally take to lead such a robust customer-service program. This saved the company nearly $2m a year in salary costs.
As the above examples show, big data is not just a buzzword, at least when it comes to enterprise software. As an investor, I’m bullish on software startups that are baking big data analytics into their products. People who use enterprise software every day on the job are simply sick of “dumb” products. If an enterprise software company is successfully applying big data to help end users be more productive and deliver better results on the job, I want to know about it.
*Battery Ventures and its affiliates have made an investment in this company. For a full list of all investments made by Battery Ventures and its affiliates, please refer to www.battery.com.
This post also ran on Wired Innovation Insights.