When measuring your online campaign's success, keep in mind that your brand has a unique set of objectives. The appropriate metrics may differ from what you expect.

Frequent visitors of online discussion boards will readily recognize the abbreviation "YMMV" -- your mileage may vary. Web commenters use YMMV when giving their opinions as a caveat to mean "your experiences may differ from mine." Internet denizens usually take YMMV to heart when reading others' comments on topics ranging from political candidates to toaster ovens. Yet marketers very often ignore YMMV when planning internet campaigns, often to their exasperation.

YMMV, of course, originates from the automotive industry. As the auto industry now lives and dies by its car models' MPGs, so internet marketing lives and dies by views, clickthroughs and conversions. More than any other medium, the internet offers the depth of measurement to track everything from the final purchase all the way back to an initial impression via a banner ad or email.

Certainly, we marketers benefit from this measurement when it comes to requesting a budget or showing success. But all too often, we get caught up in measurement and maybe even lose our way by measuring the wrong things. More to the point, we often forget that our mileage does in fact vary.

Let's begin with a typical example, something I face on a daily basis. We recommend a particular tactic to a client, anything from a simple change in the design of an email header to something as complex as a six-figure customer acquisition plan. Our client, rightly, will first ask, "What's this going to do for me?" Naturally, we won't really know until we try, but a client spending his or her organization's money wants some sort of assurance that the expenditure will be justified. So we rely on our experience and provide results from a comparable situation.

Here's where it starts to get tricky.

Sometimes we have experience in the client's industry, or we can cadge a direct competitor's experience from a published case study. But other times, we have to rely on experiences from another category. Still other times, we're trying something for the first time, which means we have nothing to go on but our judgment. While some of us rush to break new ground in online marketing tactics, many of us know full well that, as the saying goes, pioneers get the arrows.

At this point in our discussions with clients, we strongly advise that YMMV. Just because it worked gangbusters for Smith-Corona does not mean that it will work as well for Remington-Rand. (For you kids out there, these were two leading typewriter brands. A typewriter was...oh, forget it.) We ask our clients for a little bit of faith that something positive will happen.

The difficulty lies not so much in YMMV, but in the very subject of measurement. It's easy to track clicks on banners or open rates in emails, but it's not always easy to track the measurement that matters. In fact, it may not be easy to determine which measurement matters at all. Thus, just as we rely on YMMV to hedge our expectations, we often rely on measurements that do not directly reflect what matters most to our businesses.

Take the case of a manufacturer who sends emails to consumers about its products. That manufacturer can determine who received that email, opened and clicked on it and what parts of the website the email recipients may have visited. But then? All too often, that manufacturer cannot determine whether or not the recipient went on to buy that product from a retailer.

Take the case of a multi-channel retailer who runs a pay-per-click search campaign. The retailer can determine whether people clicked on the purchased search terms and maybe even identify past purchasers through judicious use of cookies or logins. But if the consumer then decides to make her purchase in the bricks-and-mortar store, the retailer will most likely not recognize that purchase as having begun online.

When discussing tactics with clients in situations like these, whose mileage matters? By that, I mean that the approach should begin by asking the question, "Which measurement matters to the brand?" quickly followed by, "What can we actually measure?" Too often, marketers accept a set of measurements as the only ones that matter either because those are what have traditionally been used or because the medium most easily produces them. Relatively few marketers take the additional step of wading through the data to find the right measurements.

For the manufacturer above, the key measurement may not be how many people clicked through to the website, but how many of them then clicked a link on the website to find a retailer and what those clicks were worth in terms of typical purchase amounts. The retailer above may wish to correlate the increase in sales of the items featured in the search campaign to search activity.

When marketers rely on the measurements that matter most, YMMV takes on a whole new cast. More than likely, the measurements that marketers develop to reflect their business needs do not match what other businesses use or what gets published in case studies. Ultimately, these measurements become proprietary benchmarks and the marketer ends up competing with himself to improve on a continual basis.

So here's my challenge to marketers: Don't settle for the least common denominator when it comes to measurement. Look into your brands. Determine what really matters and make that your key measurement. I guarantee that by keeping your eye on the measurements that can most change your business, you will achieve greater success than you could by relying on standard measurements. And in that respect, your mileage will not vary.

Chris Marriott is general manager, NY and London, for Acxiom Digital.