15 Tips from a Firm That Took Software Work to China


When the CIO for a ticketing services company heard from his CEO that they were going to scrap plans to get work done by a big-name firm in India and take it to China instead, he was a major skeptic of the decision. A year later he\’s changed his mind 180 degrees. This case study shares 15 tips gleaned from his experiences.


Paciolan competes with Ticketmaster, but with a different approach. Whereas Ticketmaster, the gorilla of the box office, typically takes over all operations of ticketing, then shares a portion of revenues with the venue, Paciolan offers reservation functionality in a self-service model. The college basketball, professional baseball, hockey, museum and performing arts centers box offices that make up the bulk of its clients license access to the software modules they want and then set up their own ticketing operations, paying a fee to Paciolan for each aspect of the service they use in their environment.


Three key factors drove the company to look outside its own IT organization for help, according to CIO Patrick Wilson. ÒThe first was our ability to scale. We\’re in a situation internally where we\’re trying to build out — rebuild an existing product, on a new technology platform. We needed the ability to scale with resources, much more effectively than we had in the past. The second key thing that we were looking to do was enhance or increase our efficiencies and quality of building products, which we were hoping to get through an offshore partner. The third was cost.Ó


The major project involves moving from a 20-plus-year-old character-based product, called Universe, which runs on IBM\’s Unix operating system, AIX, to a Sun Microsystems J2EE Web-based platform running on Linux, using IBM\’s WebSphere as the application server and DB2 as the database. As Wilson said, ÒIt\’s a significant shift.Ó


Paciolan CIO Pat WilsonOnly about half of the 50-person internal IT team is conversant with the new set of tools and technologies. Of the other half, a couple of have been cross-trained on the platform, but that was taking place at a Òlow volume, low riskÓ pace.


Paciolan\’s only previous experience with IT outsourcing was with a US-based firm that handled quality assurance and testing.


–> TIP #1. Be ready to walk away from months of effort when a new idea comes along.


That inexperience biased the CIO toward going with an India-based service provider. ÒI was more inclined to [work] a much larger organization — an India-based organization that had been there and done that, because we didn\’t have the in house expertise,Ó said Wilson. ÒWe were looking for a larger partner that could bring a lot to the table.Ó


That drove the selection process for his team. They\’d gone through a formal request for information and request for proposal with six Indian firms. Of those, they\’d selected two to go through due diligence with. They were even scheduled to visit the two companies in India during December 2004 as well as meet with potential staff members and reference clients.


Then a Paciolan board member who was a venture capitalist suggested the company look at working with Objectiva Software Solutions, a service provider with its primary technical operations in Beijing, China.


That suggestion led then-Paciolan CEO John Hnanicek to postpone the selection process, generating dissatisfaction among the IT team in charge of the effort.


ÒWe\’d spent six months making a decision and we were ready to start to execute against that decision, and now we were stopping again to go reevaluate,Ó recalls Wilson. ÒSo there was a little frustration, especially because we\’d gone through a lot of work on our part — and [put] a lot of companies through a lot of work to get to the point that we were at. Relationships had been established. All the Is had been dotted and Ts crossed.Ó


–> TIP #2. Fast-track due diligence is a reasonable route to vendor selection when you really understand what the key factors are for your project.


Those six months of vendor evaluation wasn\’t a total waste for Wilson and his selection team. It enabled them to home in on the key things they wanted to accomplish with their partner.


Wilson said he didn\’t execute a formal RFP with Objectiva. He and a manager who\’s no longer with the company met with members of Objectiva in their Carlsbad, CA offices to discuss crucial areas the client wanted to accomplish. Among topics of discussion: company size, what challenges each firm was facing from a growth perspective, how aggressively Objectiva would go after Paciolan\’s business and how openly they\’d partner.


Satisfied, the Paciolan team interviewed a series of Objectiva client references. Then they went to China and met with the principal individuals Objectiva was proposing for the project, as well as company executives based there.


Eventually, an agreement was struck. Paciolan retained the right to pick resources. Objectiva offered a Òvery effective pricing structureÓ vs. what [the client] was seeing from the Indian companies. And the service provider went about demonstrating through a pilot project how committed it was to working with the client.


–> TIP #3. A pilot project needs to be substantial enough to give you a solid sense of what it will be like to work with the service provider long-term.


Paciolan\’s pilot started around the end of February 2005 and lasted until August — about six months. Two Objectiva staff members came from China to the Paciolan site for a couple of months of training.


According to Pete Hallock, a senior engineer at Paciolan, the pilot was to evaluate how well the Objectiva team could understand the application logic, how quickly they could come up to speed and deliver against task.


Paciolan Senior Engineer Pete HallockPaciolan had a lot of tasks to complete, so it picked a small subset of the list for the China team to work on. ÒWe had a deliverable that was due near that timeframe. So we picked a couple of enhancements out of that and had them do the coding,Ó said Hallock. They also had the offshore group fix bugs in the current system to introduce them to the software.


–> Tip #4. Use the pilot project to evaluate other aspects of the company too.


Since Paciolan expected to need about 50 people, one area of concern was how well Objectiva could scale. That required Hallock to travel to China. ÒWe were trying to evaluate their HR department,Ó he said. ÒCould they bring the people in when we needed them?Ó


During that visit Hallock evaluated Objectiva\’s security procedures as well as its network set up.


–> TIP #5. Expect to run into problems during the pilot. That\’s what it\’s for.


In the case of Paciolan, Hallock reported network latency issues between the two sites. Objectiva immediately set up a Sprint backbone line as a direct link between the two companies with a private VPN to resolve the weakness. As Hallock explained, ÒThey\’re a growing company… They had too many people on their network, so they upgraded the network.Ó


At the end of the pilot project, he said, ÒThey had met our expectations. So we moved forward.Ó


–> TIP #6. An optimal offshoring structure involves domestic staff and offshore staff.


Objectiva has a technical director, Fred Nicholson, based in the Carlsbad offices, who handles communication between the two companies — including bridging the time gap. He works in the later hours on the West coast to be able to communicate with the team in China.


Another Objectiva person — Jeanne Beyer — is involved from a staffing standpoint. And Tao Ye, president and COO of Objectiva China, addressed problems that cropped up during the pilot project.


–> TIP #7. The more senior the personnel, the better their English skills should be.


Wilson expected the quality of the Chinese staff English skills to Òkill the whole deal.Ó Instead, he found himself, Òvery surprisedÓ at the quality of their English. ÒWhen we went over and met with a lot of different folks, the ability to understand what I was talking about — how I was describing a situation in our company — was demonstrated in their questions back to me… When I started to walk through our business model and our technology, to a team of about 15 people, I was expecting a lot of blank stares and a lot of silence. We ended up talking for two to three hours.Ó


Objectiva keeps an English teacher on staff to train the more senior people to learn English. Hallock said that usually once a week, during work hours, these people attend English class for an hour.


Although writing skills — particularly through email and instant messenger — tend to be uniformly good, Hallock found that among Òintermediate developers — some of the guys who aren\’t expected to do as much technical discussion and technical writing — their English skills aren\’t fluent, but they\’re decent.Ó


In the course of interviewing prospective new hires, he performed the interviews in English with an interpreter at this side. Out of interviews for 14 positions, the interpreter Òprobably got involved one or two times to make sure they understood the question.Ó


–> TIP #8. Steady state means internalizing the offshore team as much as possible into ordinary operations.


Paciolan currently has five resources internally dedicated to the project and 10 in China, so it simply divvies up the work among the entire crew.


Eventually, the entire project will involve a series of teams structured around functional areas. Each functional area will build a part of the product suite. For example, one module is a Web-based sales channel solution, which provides the tools that allow end users to go to the Web site and purchase tickets. All aspects of that module — registering, managing an account, buying tickets — would be handled by a single functional team.


The goal is to have a cross-functional, cross-continent organization.


From the Paciolan side, each team will have a senior engineer leading it; a business analyst to do functional design and interfacing with the client; a user interface design person; and one or two quality assurance analysts. Paciolan\’s people will retain UI design, application architecture, testing, and other client-facing activities. Objectiva will handle the Òactual manufacturing, the engineering,Ó said Hallock. The core development team will be in China writing code and dong some of the technical design work.


In total, each team will have about 10 people, though several will work on multiple teams. About 25 people will be US-based and 50 or more will be in China.


–> TIP #9. Managing the offshore process is a daily endeavor.


Hallock said he and Objectiva\’s onsite coordinator have a daily call with the China team at 5:30 p.m. Pacific time. Beijing is 16 hours later, so that\’s 9:30 a.m. (The China team works from 9 to 6 unless they\’re asked by the client to come in earlier.) The meeting is over the speaker phone and lasts about 15 minutes. The kinds of questions they address are, what are you working on and do you have any issues that need to be resolved?


At the end of each workday in China, the team sends an email to the onsite coordinator recapping any questions or issues that have surfaced, such as a requirement they didn\’t understand or a problem cropping up in their system. That gives the US team the day to resolve those issues.


As the project expands and more people are added, the same structure will be followed.


Person to person contact also goes on. Through their US visits (seven of 14 Objectiva team members have been to the US for extended periods), China team members know who to contact on the development and business analyst staffs in the US to answer particular kinds of questions. So they use email, as well as Skype for voice over IP phone conversations.


Hallock considers the scheduled conversations a vital key to the project\’s success. ÒIt\’s a little more organic when everybody\’s local. You can walk over to the guy\’s desk, ask him a question. You send him an email, you talk in the hall. When it\’s a team that\’s offshore, you have to schedule some time every day to make sure everybody\’s talking to each other.Ó


–> TIP #10. Expect the offshore team to show more consistency in their results. But don\’t interpret that to mean you can eliminate the onshore team.


The advantage the offshore team enjoys is that it doesn\’t get distracted. ÒThey\’re not doing anything but what they\’re supposed to be doing,Ó Hallock said. The onshore team, he pointed out, is Ògetting pulled for production support; someone\’s asking them, ÔCan you check this out? It\’s not working.\’ The people onshore have a hard time staying totally focused on the given tasks, because there are so many factors when you\’re here and the whole rest of the company is here.Ó


–> TIP #11. Don\’t underestimate the amount of training your offshore people will need.


Until recently, China didn\’t have credit cards. Chinese people couldn\’t buy anything on the Internet. Sporting and concerts aren\’t as big in China as they are in the US. In other words, the business Paciolan was in was a foreign concept to the Objectiva team in China.


That means Paciolan is constantly helping them to understand the Òvocabulary of the ticketing business,Ó said Wilson.


–> TIP #12. Even a fixed price agreement needs flexibility built in.


As Wilson explained, the contract with Objectiva is Òfixed price but there\’s a time and materials structure to the fixed price.Ó He said the two companies have come up with a monthly rate that breaks down to an hourly rate for each type of hire. Objectiva has given an assurance that so many hours will be worked in a month. Plus, there are assurances that when schedules are committed to on both sides, regardless of the amount of time it takes to reach a particular goal, Objectiva will make the date.


Although Wilson didn\’t divulge the size of the contract, which will last from 18 to 24 months, he did call it Òcost effective.Ó ÒLow is not the word I would use,Ó he said. ÒBut it\’s very cost effective.Ó


–> TIP #13. How do you know you\’re in a true partnership? The signs are pretty obvious.


Wilson points to two clues in his case. First, every time Òwe put a challenge in front of them, they came to the table with a solution.Ó Second, ÒI don\’t get them coming to us with a bill every time we do some work together.Ó He believes partners invest in working with each other, not just look for new ways to get more money.


Ultimately, he said, ÒTo a certain degree, as you sit down and look at it, our failure or success is going to be based on the relationship we have with them, and our ability to work with them and manage that relationship.Ó


Sometimes, a partnership can prove risky. At one point, Objectiva\’s technical director actually came to Hallock and inform