This is a response that I drafted to a request from an American Marketing Association member on how to setup an Internet Marketing division within his company. Some potentially valuable stuff here that I thought I'd share.

1. Develop rock solid goals and objectives for your team and ensure that everything you do under the guise of your Internet Marketing banner is done to explicitly serve one or more of the objectives. Examples of some of the goals of companies I have worked with:
- Develop leads for field sales
- Make our company "easier to do business with"
- Retain customers through e-mail marketing and loyalty programs
- Drive incremental revenue through web channels
- Close sales of a certain $$ through web channels to minimize transaction cost

2. You and your senior management need to discuss, in detail, your expectations of what Internet Marketing is to bring to the organization.

3. Additionally, a vision for success needs to be established. How will you know when you've 'arrived' or that you're even on the right track. I know, this is basic, but I can't tell you how many quandaries I've seen due to a simple lack of a shared or at least mutually agreed upon vision of success for an e-marketing initiative.

4. A common language of what constitutes a lead, inquiry, conversion, ROMI, and other web metrics needs to be established, and benchmarks need to be setup for each metric.
(here's a link to a great e-marketing tracking spreadsheet http://www.danavan.net/weblog/archives/emarketing_tracking_spreadsheets.html)

5. Selling the initiative *laterally* is sometimes one of the hardest things to do in an organization. Not knowing the specific structure of your organization, I can only speak generally on this topic, but at one of my former companies, we conducted full-scale presentations to every single department in the company, explaining the mission of our e-marketing program, the benefits to the company and the employees, and related to each department how they fit into the e-business puzzle.

6. Develop a quarterly action plan for your division. The e-marketing space still moves pretty quickly and you need to have an action plan which you can throw new ideas against to see if they meet your mission. There are things like loyalty programs and never fit in certain organizations I've worked with, that were blockbusters in others. You need to be able to evaluate every element of the e-marketing mix for feasibility in your own environment.

7. Rely on your analytics. If you do nothing else, track the effects of your e-marketing initiative. It's appalling how many companies still charge headlong into this space without a proper "Plan-Do-See" model for evaluating successes and failures and adjusting accordingly. Depending on how important this initiative is going to be to your company, it pays to develop the proper models to be able to forecast what will and won't work within your environment. Moreover, if you are diligent about your analytics, you will have an upper hand on marketing vendors who come calling making outrageous claims of improving ROI. In a former role, we saved over $100,000 in online media just by using the numbers we collected from our garden variety analytics to refute vendors' inflated traffic claims and negotiate more agreeable terms.

8. Know when to outsource. Things like Pay-Per-Click campaigns can be a sinkhole of time and e-mail marketing can consume small e-marketing departments. With each turn, it pays to evaluate outside assistance to augment your internal efforts. Always know what your time is worth.

9. How will the Internet effect other channels? Channel conflict has brought many a budding e-commerce program to its knees. Know how your channels will benefit from your e-marketing and e-commerce efforts and be able to articulate those benefits to channel partners, your internal sales force, VARS, and anyone else who interfaces in the distribution and sales of your products/services.

10. Do your homework. Subscribe to the right e-newsletters, get involved in the AMA Internet SIG, and keep your eyes open for new ideas. While much has been done, there is still much innovation left in this space. Be at the forefront and keep ahead of your competitors by devoting a certain percentage of your time to online R&D. Experiment with different models and tactics. For example - we tried an "RFI (request for information)" model of acquisition vs. our standard "RFQ (request for quotation)" model for a period of 2 weeks once, and found that 25% less people converted on the RFI model. Keep experimenting for decimal points. They all add up at the end of the year.