How can we anticipate what delivers a very poor return on investment? It's really important to determine what you should spend money on — what you invest in to be the good or service that you are — but it's also extremely important to think about what you shouldn't do.

Because many of you run family businesses, I'm going to begin my definition of strategy with an example that came right out of family interaction. My 14-year-old daughter came into my room one night holding my car keys. She told me we had to go buy glitter for a school project. I asked her why she needed glitter, and her explanation convinced me this was a great example of not-strategy — I would define strategy as a contrast to what happens here. She told me her English teacher assigned her to make a puppet to represent a Macbeth character. She didn't have confidence in her artistic skills, so she was going to give it a little something extra with the glitter. I'm going to jump to the end. It's parents night, and we see the work of some really talented kids. One of them has created a puppet masterpiece, and it's not surprising that it gets an A. The question is, what does the worst puppet in the class get?

The worst puppet got a B+, and my daughter’s six hours of investment in creating her puppet earned an A-. That is a terrible return on investment. Six hours’ time to go from a B+ to an A- is a pretty bad return on investment when you consider how scarce time is.

My daughter’s puppet was 15% of her grade, but there was almost no variance around those grades, so it didn't provide an opportunity to be above average.

Think about a good or service you sell. That good or is service is perceived along a finite number of attributes. Once your practice is being considered by a prospective client, there's maybe four things or five things left that they actually consider. Once you know what those four or five things are, you want to try to win, and you want to do it at a cost that yields us unit profitability, which means if something is high weight and has a lot of spread, you should really invest in that. If something is high weight and relatively low spread, you should try to do the minimum. We try not to overshoot the mark.

Once I'm being considered, what are the four or five things that potential clients pay attention to? In attributes that have low spread, meaning the client doesn’t really sense much, what does it mean to just meet the bar? That will inform how much work to put in.

In things with very high variance and weight, the question is what is the rubric? What do we actually use to decide something is above average? You should be looking for a good or service that a subset of the market — the bigger the share, the better — has all the attributes your potential clients are looking for. You want to focus on the attributes that the more the client gets, the happier they are.

Providing these attributes that a lot of client are looking for requires time and money. If you can make something better, and do it quickly and without much incremental spend, that's the kind of thing that provokes an arms race. What you're looking for is a way to be far above average, and to have invested significant time and resources to get to that point. If you’ve done that, your offering likely has barriers to entry, such as legal barriers. It's got some kind of increasing return advantage, and the time you've put in makes it really tough for your competitors to close the gap.

Sometimes we don’t select the right attributes to focus on. A product or service that fails make some combination of four mistakes:

Over-investing in attributes that have low weight. This is a danger when you do a focus group in your head. You develop the offering you want, and you really feel like you're on point. You over-invest, you drive down the margin or you lose unit cost. It isn’t successful.
Over-invest in product attributes that exhibit low perceived variance. You might put in too much time or resources to offer or improve an attribute that is important, but not important to the extent you invested in it.
Under-invest in attributes with high weight.
Under-invest in attributes with high perceived variance.

Some combination of these four things causes unit profits to be low and causes aggregate profits to be low.

A smart strategy for your business is figuring out your targets' most highly weighted attributes. Which attributes are table stakes? And even though we know the mantra “Do the best or do better,” that mantra has to be ignored because it's so easy to blow through money over-serving. Don’t do more than you could recoup because that attribute isn't where the game is played.

Zero in on the attributes where your client really perceives difference — where they really feel it. Determine the most efficient means to produce highly weighted attributes. Meaning, invest heavily where there's high weight, high variants and really go lean where there's high weight, low variants.

When you’ve successfully determined where to develop your offering, execution absolutely matters. Marketing is about making a connection and telling people what you are offering, and matching the message with the right people.

Marketing can’t fix a poor strategy, though. If you're executing without a strategy, you're in a continuous arms race that leads to compressed margins.

Sonia Marciano is a clinical full professor of management and organizations at New York University Stern School of Business. She teaches growth-focused business strategy, and business and marketing strategy at open enrollment programs for NYU Stern Executive Education.