Listening to social media improves your social marketing.

That idea misses the point, according to Forrester analyst Allison Smith in a recent report, “Four Social Intelligence Myths That Keep You from Success.”

Social media listening and analysis can help improve businesses in many ways, she emphasized, including better customer service, product design, or legal compliance. But Smith told me she keeps coming across four attitudes that are either flat-out wrong or are missing the larger picture.

Myth No. 1: The best way to determine the quality of a listening technology is by looking at sentiment accuracy.

Social listening vendors are not reliable sources for the accuracy of their product’s sentiment analysis, said Smith, in large part because it’s still a tricky thing.

“Sentiment analysis still has a long way to go,” VB Insight’s director of marketing technology Stewart Rogers noted in his recent report, “Social media management: Tools, tactics … and how to win.”

He pointed to a possible post like, “I want a cheeseburger so bad!” A natural language processing engine used in sentiment analysis might categorize this as a negative expression, when in fact it’s not.

Better to determine the quality of listening technology by judging its integration with your current tools like CRMs, Smith contended, and by references from customers like you.

Myth No. 2: The social metrics that we (the vendor and the marketer) track are the right ones because they are designed for the social channel.

Marketers too often track the metrics offered by the vendor, such as sentiment, reach, engagement, and impressions. But Smith said the trick is to think of social listening as a tactic, not as an initiative requiring success metrics, just as talking to a customer in a store is a useful tactic that can help informs products and services.

Myth No. 3: Listening to social improves social marketing.

Social data does a lot more than improve social marketing. It improves businesses, because marketers can better understand the customer. Smith gives the example of membership buying service DirectBuy, which is testing pricing options in social ads before taking the winning combo into the field.

Myth No. 4: Our problem is our tool.

No, she said — the real problem is the business problem has not been clearly defined, or that understanding insights from social data was not embedded as a goal throughout the company.