First Metrics

For some reason this month turned out the highest traffic in the history of this blog. It is particular interesting considering that I have not been doing a good job keeping it updated. So I feel obligated to add at least one post before December.

Let me cover some progress I made with my offshore provider I inherited through the M&A process. In particular let me cover some metrics we have established in order to control the relationship. This set of metrics is a preliminary set based primarily on what my vendor could easily retrieve from the tracking systems they already have in place.

A couple months ago we agreed on several metrics and established high level benchmarks. I believe that you can not enforce benchmarks on day one. You should setup goals and see whether they are reasonable to achieve considering multiple factors ranging from abilities of vendor and your own organization to quality of the tools that are used to collect the info.

Here are the metrics we have collected so far:

Continue reading

People Factor

I do not know how many times one of my managers or I said something like “Darn employees… Can’t live with them, can’t live without them either…” Almost all issues one faces in management career come from employees, well, very much all the issues are solved by employees as well. Every time when you deal with yet another personnel issue you throw your hands in air asking why it can’t be simple at least once in a while. Well whether you want it or not Murphy rules and if everything appears going well that only means that you are missing something.

For some reason (premonition?) I was thinking about this the Saturday before Thanksgiving week. The thought was so strong and persistent that I decided to sit down with my notebook and take a quick walk through the list of ongoing projects and open issues. The timing, from some twisted standpoint, was perfect – approaching long weekend, many people on vacation traveling, end of month, oncoming deadline for a number of high visibility projects. Nope, neither the list of projects nor the list of open issues, nor any other list I looked over gave me any reason for extra concern. Relived to the point of complacency I went on with my usual weekend chores.

Continue reading

Pragmatic Outsourcer, v. 1.0

The first volume of Pragmatic Outsourcer – the book primarily based on the materials of this blog is almost ready. There still a few things to clean up, pictures to pick, codes to request but I can see the light in the end of tunnel. The volume is going to be published as e-book and available as free download from this site.

The first volume is dedicated to the first step in outsourcing – making the decision – and that’s probably the most important step. The word “decision” comes from Latin “” meaning to cut. To make the decision means to cut other options. Many people do not look at decisions in such dramatic manner, in a way “let’s get married and if it doesn’t work out we’ll divorce”. Well, that’s a valid perspective, at least for some people. Of course any decision, even easily reversible has its consequences in many cases making reversibility quite expensive. Offshore decision is one of those thus one should chose wisely…

Choose well, your choice is brief, and yet endless. [Goethe]

I believe book format works better for delivery and consumption of material that is covered in this blog, yet it has its issues as well. In some way e-book is a static snapshot of one’s thought process while blog is a near real time stream. So both forms are important and would bring a better set of tools to the reader. That’s why I decided to take the effort to create the book.

Creating the book turned out to be harder than I thought even though a large portion of content was readily available. When you deliver material one random chapter you do not need to worry about blind spots (important material not covered), structure, consistency, etc. All that becomes a serious concern and takes substantial time to address.

With my ESL handicap I also have to deal with challenges of style, grammar, etc. that for some reason I decided to take much more serious than in this blog. I guess having ISBN number assigned to the written word raises the bar, at least in my mind. In that arena I had to seek professional help which naturally came from offshore. Eat your own dog food so to say.

Anyway, I am pretty excited and hope to put the first volume in front of you shortly and then immediately proceed to the second one – vendor selection. Hopefully in not so distant future the full scope of offshore outsourcing from decision to termination will be covered in the series of e-books I have in mind. And then… if the traffic to e-book justifies it I would love to put all my Tips, Tricks and Traps of IT Outsourcing in a hard or even better paper cover… for dummies or unleashed series… ha.. judging by what it takes to create e-book publishing in-print must be a very tough journey, so the traffic must be really good to justify it.

Invisibility Cloak of MSA

A Master Service Agreement (MSA) is intended to create a contractual framework for relationships between parties involved. Unfortunately way too often MSAs are used to protect intentional incompliance with a spirit of the agreement. When MSA is written and negotiated the parties bring to the table their knowledge of the domain, in this case offshore outsourcing services. The party more experienced in the space can predict certain behaviors and relationship patterns and appropriately protect themselves from liabilities they bring. More so that party can take advantage of less experienced negotiating partner and create an invisible cloak that will be used to hide issues and drive higher profit from the contract.

I am afraid that sounds very theoretical, vogue and convoluted… Let me suggest a couple of examples:

  • As a service provider I know that customer is likely to be late on their deliverables and my team would be spinning wheels waiting on those deliverables. To protect myself from that potentially serious issue I will put a clause in MSA that would state that if I am waiting on the customer I am still getting paid. That’s just fair, isn’t it? Now, consider what I can do during negotiations – I can downplay the probability of customer delays (most likely using customer’s ego) and shape that clause in a manner that gives me a lot of flexibility. Then, when the opportunity presents itself I can induce waiting period and rip the benefits that already embedded in the MSA.
  • Another, probably most common area, is related to provider dealing with the resources on their side. There are many areas where supplier can negotiate “reasonable” terms that have nothing to do with reality of the situation. For example, if a software developer quits another developer would be put in his/her place and ramp up period should be the industry’s standard 2 weeks. Industry standard? When I bring onboard a new developer it takes 2-3 month for him / her to become fully productive how come it takes four times less with an offshore guy? That’s not the point though, no matter how many weeks of shadowing you might negotiate the realities of delivery against the item in MSA remain practically unknown, and thus could be manipulated to fit provider’s objectives.
  • Even a very straightforward items like “body count” becomes pretty vogue and unenforceable. Imagine that you are trying to count people in organization and people always move from one office to another. Getting the numbers right would be quite challenging. Just a few weeks ago i spend almost a week trying to figure out how many QA engineers I have on staff with my Indian offshore operations. The numbers varied greatly depending on who I’d ask. Most precise figures came from the vendor, in that light resorting to MSA as a lifesaver is only natural. Yet, if you think that if my development manager thinks that there are 2 QA engineers on his project while my provider tells me that there are 5, something is seriously wrong here. I bet it means that I get the work of 2 while paying for 5…

In general what makes an MSA an invisibility cloak is not bad intentions of the vendor, but buyer’s inability or lack of desire to enforce it by staying on the top of engagement. If you do not control the deliverables each step along the way, if you do not verify timesheets and assignments, if you hope that the MSA will prevent me from issues and problems of malicious or delinquent nature you will most likely fail. In that case the MSA will become opaque and impenetrable defense mechanism for the vendor. I guess Invisibility is in the eye of the beholder.

Steps to making an MSA transparent are obvious – focus on execution, control of deliverables, etc. Considering an example of team turnover. A realistic ramp up for a developer in terms of productivity would be 25% first month, 50% second, 75% third and 100% from that point on. In that case over 12 months developer produce 1050% of the monthly allocation. Suppose a developer quits after 6 months and spends one month training a shadow resource (it’s reasonable to assume that that between two of them productivity for that month is 100%). In that case total productivity over the year will be 975% or ~7% less. If we have two replacements over the year the figures would be 900% or ~14% loss of productivity.

That could be easily translated to the rate impact – if your rate for the developer was negotiated at $25 per hour in the second case you paid roughly $27 and $29 in the third. Of course not controlling these figures makes the difference invisible… The magic spell to make the cloak transparent would include linking turnover baseline to rate and more important watching it over the case of the engagement.