Pleasing the Customer(s)

When I worked in the mobile devices division at Microsoft we had an ongoing discussion about who our customer was for our mobile device offerings:

  • The carrier (e.g. Sprint, Verizon, AT&T, etc.) upon whose network the device would run.

  • The OEM who would make the hardware (Dell, HP, Motorola, etc.).

  • The kid who would use it.

  • The parent who would buy it for the kid.

  • The department(s) within Microsoft which might profit from services sold through the device.

  • The random executive or product manager who had an opinion of what features should be on the device.

  • And of course: ourselves!


So which are the customers? Answer: All.

Customer: Anyone who lays a legitimate claim to directing your software development work.

Imagine how difficult it is to develop requirements for any one of the "customers" listed above. Now imagine creating requirements which balance all of the needs of the competing interests. Ugh.

We face a similar, though not so extreme, challenge in I.T. We have internal customers (typically product managers) who work for the departments we serve. They have their own management chain who have opinions of what work should be done and how. We have our own management chain which has its opinions. And then, of course, we have the end-users for whom we're collectively building the software. Pile on top the fact that "knowing what one wants in the software" is not an inherent skill. It's something most people think they do well, but don't.

So what to do? Here are some suggestions:

  • Start with high level requirements. Make sure you understand what the goal of the project is. Stand firm on this point, and don't relent for any amount of begging, threatening or bribery. Get the value proposition documented in plain English and make sure your customers, all up and down their management chain, agree. Treat this with the utmost importance and respect. It will become your rock later on when disconnects occur between product managers and their management. The value isn't so much in having a tool to cover yourself. Rather, documented requirements help you help the customers get on the same page about what is expected.

  • Always be an advocate for the end-user. It's your solemn duty. Whomever your customer is, they care about the end-user, even if they don't understand how to delight them. Use whatever tools you choose or have available: focus groups, usability studies, contextual inquiry, intuition, whatever. Just make sure you understand the user(s) of the system and that you champion them at every turn. Your customers will thank you (though maybe later).

  • Use functional prototypes to narrow the "gap of misunderstanding" up front and get everyone on the same page about what is to be built.

  • Prioritize your customers. Understand which is most important to satisfy and take care of the high priority features for the high priority customers first.

  • "Just say no" to executive/management pet features. Executives appreciate those who "speak up." If someone in your management chain tries to get you to add or subtract features and you know in your toes that its the wrong direction, then push back! If you've said your piece with all vigor of heart and the answer is still no, then press forward--you've done the right thing by speaking up. If having done so becomes a "career limiting move" then you may have the wrong management anyway. I speak from experience. The people in our organization typically take great pleasure in pushing back on me.


Having multiple customers can be a great challenge, but when you mostly satisfy most of them (in priority order) you accomplish something which brings great satisfaction.

5 comments:

  1. The customer also gets more difficult to please or understand when they come from different countries that have different cultures, languages, and values. Globalization in the scale that it is today adds even more complexity to the "Customer" perspective.

    ReplyDelete
  2. Good insights! Would you consider extending the first bullet from "Start with high level requirements" to "Start with a vision and high level requirements"? My team has struggled too frequently with absence of the "biggest picture" even when they had high level requirements and detailed requirements (user stories).

    The "Just say no" bullet is more difficult for me, since "pet features" seem like they should compete against all the other requirements, and that competition in the marketplace of ideas should bring the winning ideas to the top. If the product owner describes the other features that will be sacrificed (or the time that will be added) to provide the "pet features", and the executive overrides the product owner's decision, then the competition has taken place and the executive has decided to become the product owner for that part of the product.

    ReplyDelete
  3. Your suggestions on balancing the competing interests of "The Customer(s)" accomplishes something very important for the IT organization as well as the delivery of a satisfying product to the user.

    Standing firm against the temptation to “just get on with the project” without a clear plan, advocating the end-user, prototyping, prioritizing and sometimes saying “no” all serve to establish IT as a partner for process change ... not a subservient order taker. When IT steps up to the role of true partner and collaborator for process change … only then can optimal effectiveness can be achieved across all the competing interests.

    ReplyDelete
  4. Correction to Typo in previous post ...

    Your suggestions on balancing the competing interests of "The Customer(s)" accomplishes something very important for the IT organization as well as the delivery of a satisfying product to the user.

    Standing firm against the temptation to “just get on with the project” without a clear plan, advocating the end-user, prototyping, prioritizing and sometimes saying “no” all serve to establish IT as a partner for process change ... not a subservient order taker. When IT steps up to the role of true partner and collaborator for process change … only then can optimal effectiveness be achieved across all the competing interests.

    ReplyDelete
  5. I have an odd comment or question but somewhat related to this post for Bro Dehlin. For Pocket PC users that have the newer microsoft pocket pc operating system what would you recommend for chuch publication downloads? Such as the scriptures, lesson books, etc...

    [Joel: I don't have a recommendation, but I'd be interested in what the community thinks... ?]

    ReplyDelete