Hi Greg,
I was enjoying your blog until I came across the following comments on Indian outsourcing:
"If we don’t move from straight-technical on-site/long term into becoming more business-specific technical consultants, then we end up in a rate-war with Indian outsourcers because a “J2EE guy” or a “.NET Guy” becomes a commodity. And once that happens, a North American company can not win. …our quality will always be better, but Indian outsourcing rates are always so compelling to upper management, that you can’t compete…we simply can’t give the client a “senior” .NET expert available for $20-$40 and hour and keep a straight face. Indian outsourcers can’t really provide this guy either, it takes their experts 8 hours to do what we can do in 1 hour…so you can pay us 1hr * $150 or then 8hr*40 ($320) to get the same job done…but it takes upper management of a client many failed projects and a long, long time to realize it."
To be honest, I felt that the comments above were a slap in the face to all of Infusion's employees who are of Indian descent. My interpretation is that Indian programmers work at 1/8th the productivity and with greater programming defects than a North American programmer simply by virtue of being Indian.
[sentenced removed because of public view] However, Infusion consultants are smart people and eager to learn so they are able to quickly become productive with new and existing technologies. There is nothing inherently North American about that ability to learn that prevents an Indian company from doing the same.
The industry bashing of Indian outsourcing companies is a small pet peeve of mine, perhaps because of my own nationality. Through my assignments with [client X], I have worked with some of the Indian outsourcers and I found that the only thing that separates them from the typical [X} employee in New York is the time zone. My former [X] manager, Krish [Y], recently went on a trip to India to evaluate a number of outsourcing companies. I was able to sit in on one of the meetings where he gave a summary of his findings and the summary of his trip was that he was impressed with the level of organization, quality, and hiring process at a number of the Indian outsourcers.
My take on the issues with Indian outsourcing is that the defect rates and lack of productivity are a result of the growing pains of outsourcing programming projects to a company in a different country that speaks a different language. Since Infusion has direct experience with trying to start up an outsourcing branch in India, I would be interested in reading your thoughts and getting more history on the Indian outsourcing attempt.
------------------------------------------------------
I think I see what is bothering you. I think it is the line: “it takes their experts 8 hours to do what we can do in 1 hour” Let me clarify:
The reason the outsourcers produce a guy who takes 8 times longer isn’t because their people are Indian, that, obviously, has nothing to do with it… it is because most outsourcers (particuarly the large ones) are not honest about the level of experience and do not ultimately deliver what they promise. As I will point out later in this post, a truly “senior” Indian resource is not inexpensive…they are close in salary to the same level of resource in Canada. Neither we nor they can provide these guys for $20 an hour. So, the Indian outsourcers instead take junior people, bill them as “senior” and/or have someone other than the senior person they show to their client actually working on the project…hence, that and other failings in their business model results in them taking 8 times longer.
With that said, let’s make a CLEAR distinction between Indian outsourcing companies like Wipro and TaTa and people of Indian descent. You can be critical of the business practices of one without implying anything about the other…the two are unrelated. I don’t like Halliburton or the behavior of big US oil/energy companies, but that doesn’t mean I am anti-American or slapping North Americans in the face. In fact, the large Indian outsourcers have locations now in Canada, Ukraine, China, Pakistan and other places. So, if I feel (and I do, *strongly*) that Indian outsourcers are not good value (for many kinds of projects) no matter *where* in the world they outsource, I am not slighting people from Canada, or people of Indian, Ukrainian, Chinese, Pakistan or other descent.
You also mention Infusion’s experience in starting its own Indian office (which we did a few years back). I’ll go into that effort at the end of this post, but the reason we went directly to India ourselves and invested in starting our own Indian subsidiary instead of partnering with an outsourcer was *precisely* because we believed Indian talent to be on-par with anywhere else. We believed their efficiencies where obstructed because of the big outsourcer/middleman in the middle and that if we went to India ourselves, we’d be able realize substantial advantages by doing it ourselves. What we found out is that, yes, the talent is on-par with everywhere else, the people are as ambitious, educated and driven…but the business model of leveraging Indian outsourcers as a huge cost savings alternative is a myth.
Lastly, let us also remember that Infusion, itself, is run day to day by someone of Indian descent, Alim. You can and should talk to Alim about his experience with the Infusion Indian office…maybe I can Alim to do a post. But the fact that we have Moslem’s, Jews, Catholics, Indians, Russians, Korean, French, Chinese…people from every part of the world and every orientation at Infusion…that is one of my biggest sources of pride and precisely what makes us strong. So, not to be tough but, your phrase, “My interpretation is that Indian programmers work at 1/8th the productivity and with greater programming defects than a North American programmer simply by virtue of being Indian.” is, as you say, your interpretation. I recommend that you not "interpret", but separate the sensitivities that you genuinely have from what I have *explicitly* said.
So, to summarize: it is Indian *outsourcers* that have 1/8th the productivity of Infusion, not people of Indian or any other descent. Here are some reasons why this is so and note that NONE of these reasons have anything to do with one’s nationality or heritage, it is solely a business/organizational matter. Consider the following:
1) The Indian economy is hot: good resources jump CONSTANTLY between different outsourcing companies. There is nothing more disruptive to a client than losing key people familiar with their environment and having to ramp up others. It is a *major* source of irritation for clients and results in frequent delays. Outsourcers try to make up for this with extra process, but net net, is that nothing is a substitute for good people who know your business. This is one of the things that make up that 1/8th productivity hit.
2) One of our largest clients is ramping down their top Indian outsourcing relationships, and they are not alone. If Indian outsourcing works so well, why would this be so?
a. One reason is that the client decided to create their own, wholly owned Indian subsidiary, which is *fine*. That might very well work great. I never said Indian development didn’t work, I said Indian outsourcing companies don’t work.
b. Another reason is that they were not able to handle complicated specifications very well. Migration projects, great. Projects with clear scope, great. A trading system, hit and miss. I appreciate that your manager has had good experiences, but I got these sentiments from the very top of the organization. No one want to be buried under the specs, processes, checks, counterchecks, etc. of trying to communicate complicated ever changing specifications to an outside, distant company. You are just better off doing it yourself in India or wherever else makes economic sense.
c. A business process typically employed by Indian outsourcers is to offer extremely low rates to start, but then increase them substantially once a dependency is established. I have seen this happen multiple times at the very highest levels of different clients.
d. Massive concerns about intellectual property: people jump, IP laws (while being worked on) are not yet in a state that major Financial firms would like.
e. Much like Krish said, any analysis or onsite visit of a large Indian Outsourcer is impressive. No doubt. But these firms are ALL about impression. They say all the right things (we are CMM certified, we have great processes,etc.) they are the best salespeople around. That is why your client singed up with them in the first place for HUGE contracts. Yet now they are ramping down. The proof is in the performance, and the performance is hit and miss. Maintenance/migration/administration work great…but sophisticated projects rarely succeed (or, if they succeed, they don’t do so with the savings promised). If the Indian Outsourcers were so terrific how, after 30 years of Indian Outsourcing companies being around is not everyone using them exclusively? IBM was using them in the 70s, they were the first. Their growing pains are quite old…they have been around as long as IT development has been around. The trend in Wall St. is relatively new. It had its peak a few years ago, now it is settling somewhat. The trend now seems to be where firms have their own, captive Indian site (and sometimes an Asian site or site elsewhere as well). That, I think, will work *much* better for them. No middleman, and the ability to freely move company resources back and forth instead of relying on a vendor’s middleman.
3) We exist. How could we exist and continue to grow when for every single service we offer, bar none, there are outsourcing companies that can offer lower rates than ours? We are growing deeper into large financial firms, every one of which have huge, Indian outsourcer arrangements. Why are we still getting projects?
4) The distance and change of time zone is a subtle but very real problem, whether people are willing to agree or not. It is exhausting for a North American based manager to be on the phone every day discussing deep technical details at 7pm or later at night with the outsourcer. These folks have families, and are not in their best mental state after a solid day of work in NY. You can say the same for the folks at work in at the Indian outsourcer: they have families too and don’t want to be awake at night and sleep during the day. So, someone is always exhausted when you talk. Specs get fuzzy, details get lost. Face-to-face and voice communication is *critical*. For all the electronic forms of communications in the world, when you are trying to figure out what an angry trader wants on his new trading system, you’d better be talking…often. And you both better be very sharp. Placing a day between a conversation and an action slows things down…hence, another part of the 1/8th. None of this is to say that the time zone can’t work in your favor, for support, it works great. At the end of the day, you give the outsourcer your changes, go home, go to sleep, come in the next day and there are your changes. But this is not complex project work, this is support/modification of existing systems.
5) Software is no longer as much “spec driven” but more “iterative”. By the time you have a trading system or some other complicated piece of software fully spec’d out so that you can ship it to a remote destination and feel comfortable that they will be able to write it, you no longer need to ship it. Technology is no longer difficult, with tools like .NET or Java/Eclipse, nothing is really that hard to put together these days. But nothing we do is really difficult technologically provided EVERY detail is nailed down before you write a line of code. But *specifications* are difficult. If you look at the kinds of projects Infusion takes on, we take on things where the specs are fluid: our skill is being able to adapt, to work iteratively, to be able to make our own decisions in places where a spec is fuzzy. If we tried to simply say to our clients, “this spec is incomplete, please be precisely clear on A,B,C and D” we would not be useful to them. The business, whom we all serve, does not read specs. The human reality of getting complicated projects done is that you have people close at hand, who understand the political realities as well as the technical. Sometimes you need a detail about the business no one knows…and no one knows who knows…you have to find the person, take them to lunch, chase down the trader (dodge a phone hurled at you if he/she is in a bad mood) and figure out that detail…THAT is what it is all about. I appreciate that not everyone may work on that kind of project, but this is the reality we live in. We don’t live in a nice, discrete, process-driven reality…we work in the supremely fluid, irrational, tumultuous world of capital markets where the first people into a new line of business with the right technology makes a killing. Try to spec that. Also note:
a. A typical technique Indian Outsourcing firms like TaTa and Wipro use when they miss deadlines or are not fulfilling spec is to trade time. They say, “no problem, sorry…we’ll fix that, no charge.” This is because their internal development costs are extremely low, so time isn’t really a problem for them. But there is a very real cost to a business when a project is delayed. I have seen this time and time again in many firms.
b. Many bright professionals in India are trained deeply in CMM and these kinds of process paradigms. Unfortunately, for the kinds of projects we take on, that education doesn’t always fit. We don’t have hyper-evolved specifications from clients that allow for these kinds of folks. And, frankly, folks deeply rooted and trained in very specific positions don’t always enjoy more the fluid environments we are forced to live in.
Now, you asked about our experience when we started our own office in India a few years back. We saw all 5 above and many other things. We basically lived as in a microcosm of what large firms experience when Indian outsourcing. That is one of the reasons I hold my opinion so strongly…I was there, I did it, I spent the money, built the business, tried to make it work, fought the battles and, ultimately, closed it down. One of the things we found out while there, was that in spite of Indian Outsourcing companies claims that “senior” people could be had for low salaries (<20k) we found that it simply wasn't true: when we got to Bangalore ourselves and opened an office, we found that while you can get junior, less experienced talent for this…top Indian talent has comparable salaries to Canada. You also have to deal with the fact that the top-experienced people in India can come to the US or go anywhere else in the world they like through the outsourcer…so they are priced for these markets not the Indian market. Good people are not a bargain no matter where you go. Thus, if you want “top” person who understands the business, has solid experience in code, is a great communicator, etc., their salaries are very close to the comparable salary of that person in Canada. So, you have two identical people: one very distant in a different time zone, and one 1hr away from all our clients. Who do you pick?
So, in the end…our experience with our own Indian division was that the talent was there, but by the time you paid what the market rate for our quality of resource, added all the other expenses (monetary and administrative) of dealing with and managing a remote entity and of moving complex projects back and forth, our analysis showed that India was, in fact, a better price point than North America…but that Canada could do the same work for actually less (all expenses and administrative items considered.) Thus, Infusion today has the bulk of its resources and development staged out of Canada. That is why we became a “near-shorer” instead of an “Indian out sourcer”. And, if you look at the trends, note that many of the Indian outsourcers are now offering near-shoring in Canada, just like us, in order to assuage the concerns of financial services firms. Why are they doing that? (Check the Globe & Mail newspaper in Canada…they have a bunch of articles on the India Outsourcing in Canada trend).