The Structure and Delivery of Small Project Consulting

When you consult on larger projects or many varied projects, the majority of the time is spent in client specific activities. For instance, you may spend a couple of hundred hours developing a rich domain model to really understand their unique business processes. However, many smaller consulting projects often have strong similarities to other projects you've undertaken (how much real innovation can you fit into a 60 hour web application?!). The question is what is the best way to handle that . . .

I do some software product line, architectural and code review consulting projects, but the vast majority of the projects I do (by volume) are custom web applications. Often they are "with a twist" sites - a commerce engine with clever discounting rules or a content management system with rich, related custom objects. A secure document exchange with a custom UI and workflow requirements or a simple "web 2.0" community site with events, content management, blogs and some kind of user feedback mechanisms.

The opportunity and the challenge with such projects is that most of them are the same. When was the last time you did a project that DIDN'T require some kind of authentication system? And if there's authentication, there is the question about whether users should enter an email address or username to log in, any password strength rules, whether the login page justifies a secure certificate and lots of other similar issues. The answers may vary from project to project, but the questions and information are pretty consistent.

Equally, when was the last time you built a project that didn't send an email to someone for some reason? But if you send emails, should you use your web server to send the emails or integrate with their SMTP servers (IT and deliverability issues abound). What about the format - HTML, text or both? For HTML emails, how wide should they be, and do you want to test them without images so you see the default view that many users will now have of your messages? And do you need to take a snapshot of your email templates using multiple email clients to see how it'll look in GMail, AOL, Outlook and Entourage? And then there is tracking emails - tracking bounces, views and click throughs and all the usual caveats about filters that eat emails without bouncing them and users that view emails without downloading images.

The challenge with a simple commerce site is that if you were to spend the time fully covering all of the issues that can arise (anyone has the zip code vs. state discussion with a client recently?), the cost of the project would treble just from the time it takes to describe all of the issues. The good news is that most clients don't care about most of the issues, but the bad news is you never know how many questions will arise or whether you've got the "client with a million questions", so you can't be sure of the cost of the project until it has been completed.

What I'm trying to figure out are some options for the delivery of common questions/information. Of course, you could just sit down with the client and burn hours. That works for me as a last resort and I know some clients are willing to pay $125/hr for you to say something they could have read on a FAQ on your website, but I prefer to have more options for my clients to get better value for money if they are open to that.

I'm going to play around with the structure of the information, but it seems to be in general terms a decision support system where you might have information and a question with an answer which could be various data types - including an enumerated list, where supplemental information could be associated to questions where the user wanted more information (or even a wizard to drive their choices) and rules could be added so that depending on the answer to one question it might add additional questions to the list. Such an interface could be used either directly by a client or by a customer service rep to handle most questions that are usually treated as cnsulting, freeing up the consultant to handle only questions for which they haven't been able to develop reliable heuristics for (I like to think of those as questions that the consultant is still dumb about - in my opinion, you don't REALLY know what you're doing as a consultant until you can teach a computer to do it for you!).

Of course, books, articles, FAQs, videos, classroom training and intake materials are all possible formats for such material, but where the clients job is primarily to know just enough to feel comfortable making decisions, a decision support system that replicates the consulting experience seems a more efficient delivery mechanism - especially for the many clients who aren't sufficiently motivated to learn about the field - they just want to answer the questions and move on. I think it would be important to also have default answers where possible so clients wouldn't even have to think about the many questions they probably don't care about and some kind of build system on the back end where the cost of a client changing their mind late in the project would be as low as possible.

What do you think? Anyone out there built decision support systems? Do you email your clients recommended reading or give them questionnaires to fill in between meetings? How do you leverage the value of your time as a consultant? Input much appreciated!!!