Solving the Recruitment Puzzle with InterviewStreet

In August 2011, InterviewStreet became the first Indian company to be chosen for Y Combinator, an incubation program that has helped launch companies like Reddit, Airbnb, and Scribd. 63 companies had made it to the list last year, a mere 3% of all applications.

Founded by Vivek Ravisankar and Harishankaran K., InterviewStreet creates customized programming tests for companies, that are used to evaluate and screen candidates for recruitment. Starting with their first customer, Bankbazaar, the list of companies that use InterviewStreet for recruitment has now grown to include the créme de la créme of the startup world – Facebook, Zynga, Groupon, and Spotify among others.

In October 2011, InterviewStreet had its first CodeSprint – a 72-hour contest to help developers showcase their skills to potential employers. The CodeSprint had 3,143 users viewing the problem. 32% of the viewers submitted a solution, and 31% of those that submitted solved the problem and were eligible to apply. Only half of those eligible actually applied. For the second CodeSprint held in January this year, the duration was cut down to 48-hours. It had 86 companies looking to recruit, and had 5,221 users viewing the problem – up 66% from their last CodeSprint. 36% turned in a solution, and more than half of these solved the problem and were eligible to apply. 66% of those eligible applied. The distribution of eligible programmers by location indicates the wide reach of InterviewStreet – 38% each in North America and Asia, and 14% in Europe. The next CodeSprint is scheduled for May 11, and the one after in August/September. In the meantime, InterviewStreet plans to hold CompanySprints (shorter duration CodeSprints run for a specific company), a SpecialitySprints (like CodeSprints, but designed for specific job types).

—

The founders, Vivek and Hari, graduated from NIT Trichy in 2008 and went on to work for Amazon and IBM respectively. InterviewStreet began as a platform for conducting mock interviews – setting up practice interview sessions for college students with ex-employees of the companies that they were interested in applying. More than a year into their jobs, both the founders quit to focus on their start-up. After 9 months of little progress, they gave up on their original track of mock interviews. (Vivek has an insightful post on why it didn’t work on his personal blog). Having recognized through this process that recruitment was more of a challenge to companies and took up considerable time and effort, they changed track and morphed InterviewStreet into a platform that helped this process.

InterviewStreet’s revenue model appears to rely on a flat-fee to companies that recruit candidates through their site. For the CodeSprints, companies are free to join and interview, but have to pay InterviewStreet a flat-fee if they recruit the candidate. According to a Forbes article from last August, InterviewStreet charges a $10,000 referral fee for every hire, though this seems like a number that may be applicable to their US clients. They are said to be profitable with revenues doubling every month, and the article above places them in the list of Y Combinator companies that will be worth a billion dollars. Current investors include Morpheus Venture Partners and Y Combinator, with a total funding of $200k. Competitors include Gild and CodeEval.

At a time when some of the successful Indian startups are adapting concepts from the West (think Amazon-Flipkart, and Groupon-SnapDeal), InterviewStreet is one of the rare ones that are truly innovative and global in their reach. While the space they operate in is bound to get competitive, we are optimistic about their success and ability to scale, or we are allowed to speculate, an acquisition by the likes of LinkedIn or another job-hiring site that may help InterviewStreet benefit from the synergies of having access to a much wider market and expand into areas beyond software recruitment.

Taking the redBus

The story of how redBus was conceived is well known by now. During Diwali in 2005, Phanindra Sama goes to his travel agent to book a bus ticket from Bangalore to Hyderabad. His agent does not have a ticket. He made a few calls. No luck. He is asked to check with another agent. Phani goes to the next agent, and then another. The same process is played out again – the multiple calls, the referral to another agent. After 6-7 visits, Phani gives up. While his friends are away with their families, Phani is stuck in Bangalore wondering if there was an agent out there that perhaps had a ticket for him.

Sharing his experiences at the TiE Pune My Story session last September, Phani appears modest; a genuine kind of modesty without affectation. He alludes to growing up in a village, being sent to his uncle at Hyderabad for school, the state ranking, the degree from BITS, and the job at Texas Instruments. That in itself would have counted as success by the expectations by most parents. Phani then decides to turn conventional wisdom on its head. With two of his friends, he believes that they can solve the bus ticketing problem  by creating a system that lets bus operators share their inventory on a common software platform, and make this information available to agents and customers looking to get tickets. That way the bus operators maximize the seats they fill, the travel agents sell more tickets, and the customer finds the ticket that would have otherwise been lost in the maze of agents equipped with incomplete information. That way, Phani would have probably found his ticket home that Diwali in 2005.

—

redBus’s revenues comes from their three products. The front end product is redBus.in, the consumer portal that lets you make reservations online or through calling up its customer service number. The two back-end products are BOSS (Bus Operator Software System) that allows bus operators to manage their available inventory, and Seat Seller that is used by travel agents to see the available inventory, commission, and booking history among others. redBus derives a transaction fee on transactions that take place within each of the above three products – when you buy a ticket at the bus operator or their website (hosted by redBus), when you buy a ticket at a travel agents, or when you buy a ticket at redBus.in. According to Phani, the three systems feed each other. More bus operators signing up lead to more consumers and agents dealing with redBus, which in turn lead more operators to sign up.

In the 2010-11 financial year, redBus had revenues of 120 crores. For the prior year their turnover had been 55 crores, and the year before that, 25 crores. Their revenues had been growing at about 120%. For the 2011-12 financial yearn that starts in March, they projected a turnover of 250 crores. By September of that year, they had crossed 200 crores, causing them to update their forecast to 400 crores. According to Phani, factors that have led to this unanticipated growth are – increase in BOSS adoption that has led to high inventory, the halo effect generated by a slew of online companies wooing customers to make online purchases, and finally word of mouth.

—

When Phani anticipated building the site, his idea was to create an open-source platform that he could give away to bus operators. Most of the operators he and his co-founders spoke to were dismissive about the software. Yeh sab airlines ke liye chaltha hain, bus ke liye nahin chaltha (this – inventory management – works for airline reservations, not for bus reservations), they were told. Despite the initial resistance from the operators, Phani and his team worked on developing an inventory management solution. They were really starting from scratch. The founders had backgrounds in VLSI and embedded systems. They bought books on .NET and databases and divided up the work. As Phani describes it, they literally started with rendering ‘hello world’ on a webpage.

Once they had the solution ready, they were hoping for the bus operators to immediately adopt and embrace it. But the bus operators kept putting it off. That’s when they found a mentoring program through TiE Bangalore. With advise from their mentors, they were able to shift focus to getting customers to buy tickets from them through blocking off inventory. In due course, they hoped the bus operators would see them as credible, and would be motivated to buy into the inventory management system with a view to getting more customers.

There were other challenges to be encountered. First was the slow growth that confronts a new venture. Four months into leaving their jobs for the startup, they realized that they were making only about Rs. 2000 a month, hardly enough to justify the pay that they had given up. But they persisted. There was also competition – rich competitors willing to throw money, computers and even personnel to help operators adopt their inventory management software, and attempting to lock the operators into exclusive contracts. The key game changer, redBus’s mentors advised them, was getting the customers – the operators would follow. And that’s what happened.

—

‘We are a bus ticketing company’, says Phani. ‘not an online bus-ticketing company’. The difference is that online companies try to change customer behavior to purchase online. redBus wants to go where their customers are, and not necessarily force their customers into their way of operation. redBus has 7 call centers, and they are opening physical locations where customers can walk-in and buy tickets. 25% of their business comes in from these call centers.

In describing what it takes for a company to scale, Phani draws on examples from education. It takes incrementally more effort to go from 80-90% than it takes to go from 70-80% in an exam. The difference, he says, is brought about by dedicated and persistent effort. He talks about this analogy in the context of conversion rate on travel websites. The global standard is 2-3%. redBus started at 1%, but through persistent effort have managed to get more than 10%. It is his equivalent of scoring more than 90% in the tests.

—

When in Class X in Hyderabad, Phani talks about how he wondered if he was having access to good teachers. What if there were better teachers in other schools that would help their students perform better than him? His uncle reassured him that even if he went to another school, he would still wonder if there was a better teacher elsewhere. Moreover, his present school was the best means he could have at his disposal. That seems to have convinced him them. But when it came to finding the elusive ticket, he was not one to rely on dubious information. He was willing to organize the entire bus ticketing industry to keep up with his expectations, and in doing so benefit millions like him who now have a convenient way to travel to their near and dear ones.

Alexa’s Top 5 Indian Sites in India

We checked to see which Indian sites show up on the top in Alexa’s site rankings for India. By Indian sites, we mean born-and-bred desi, so no Google India domains, no Facebook, no Youtube.

At #11 Indiatimes is the top dog. Click on their url, then go make yourself a cup of chai, and you will still be back in time to see the page load. But hey, they are #141 on Alexa’s global list. Who are we to say anything?

Rediff comes in at #12. Cleaner interface, and their list of 5 headlines has that perfect blend of news, celebrity photo, cricket, celebrity photo 2, and a random topic that you’ll probably never get down to reading. Global rank: #123

Relief at #17 in the form of ESPN Cricinfo. Finally a site that does one thing, and does it well – extensive cricket coverage and stats down to the domestic matches. A well deserved global rank of 185.

At #19 – in.com. Let’s you create @in.com email ids which, in this day and age of gmail and growing mailbox sizes is still worth a brag. Their interface makes Indiatimes look like it won a design competition. Global rank of #342

Fifth on our list and coming in at #22 is HDFC. Yes, the bank. At a global rank of 467, that’s some achievement for a banking site. Also, the only one among the five without sponsored ads.

SnapDeal and the Coupon Mania

How big is the online coupon market in India?

If you had ventured to find out more than 2 years ago, measurably not much. In November of 2010, comScore estimated 10.5 lac (1.05 million) visits to coupon sites in India. In November 2011, this number had grown an astounding 629 percent to 76 lac  (7.6 milion) visits. 67% of this was snapped up by Snapdeal. In hindsight, an amorphous market for online coupons seems to have existed all along. It took an enterprising startup for this to morph into something tangible and measurable.

Founded in Feb 2010 by Kunal Bahl and Rohit Bansal, Snapdeal has in its 2-year span achieved more feathers than can fit a cap. It is the 25th most visited site in India according to Alexa. It has received 50 lac page views last November, and was valued at Rs. 1,000 crores last July. Heck, it even has a village named after it. With its hold on 50 cities and 30,000 merchants, Snapdeal has become a formidable presence in the online coupon marketplace where the barriers to entry are low, but the barriers to dominance are now significant. The next most visited site, mydala.com, had only about 16.7% of the total visits last November – modest, but not what you would call a close second.

There is no precedence in India for what Snapdeal has started here. If one looks beyond, there is the overarching presence of Groupon – the Chicago-based mother lode of online coupon sites. Snapdeal’s likeness to Groupon does not end with the uncanny similarities in their web interfaces. They have both had similar experiences – a meteoric rise within a short span, deal-hungry customers, ecstatic merchants, high valuations. But with one difference: Groupon is starting to experience the waning edge of the curve, or what some might call a spike – a slow decline in euphoria, disgruntled merchants that have not seen commensurate benefits for the discounts they offer, their own accounting fudge-up close to their initial public offering. These have cast doubts on the future of the company that has been the darling of internet stocks in the Valley, with one Forrester research analyst calling it ‘a disaster’ and ‘a shill that is going to be exposed pretty soon’. Snapdeal hasn’t experienced this. They may learn from the missteps of Groupon, and figure a way to keep their users, merchants, and investors happy. Or they may ride the wave until the novelty wears off at the end of which they may be left at a less euphoric but more stable state of existence. Co-founder Bahl expects the company to go from $35 million to $100 million in sales by the end of the next financial year, and cross $200 million the year after. It will be interesting to see what the comScore numbers on coupon site visits will say this November.