CIO — In most product-driven firms, product planning is one of the highest leverage processes in the whole company. There's a huge difference in the profitability of a "hit" vs. a me-too product, and a dud is worse than just unprofitable. By its nature, product strategy is as much art as science, but bringing more hard data into the process improves the quality of prioritization decisions.

Product strategy needs to be a mix of engineering/operations plan and market survey, but most market survey techniques are quite vulnerable to big procedural and statistical problems. The iconic bad product of the 1950s was the Edsel, yet it was the result of the most thorough surveying processes of its era.

Fortunately, CRM systems naturally contain information about products, customers, and features. Further, modern CRM systems store a sequence of interactions that make the data support richer inferences about "what's important to customers."

• The most essential data for product planning is transactional: what products sell where, at what discount level, to whom. While the core of this data may be available from your accounting or order-entry system, the CRM system adds color to the transaction, such as the vertical market, the names and titles of the (likely) users, and the length / complexity of the sales cycle. From this data, you can understand which products tend to be bought together, what are the messages, campaigns or offers that tend to stimulate sales, and which competitive situations are the most favorable. From this, you can infer which feature improvements can help you win more often.

• The other essential data comes from customer service interactions: what products and features seem to cause the most trouble. If you've got a customer support portal set up, you'll be able to identify the cases, solutions, and knowledge-base documents that are most popular (where "popular" means "common problem area"). If your customer support team logs their hours against cases and bugs, you'll also be able identify the most expensive product areas to support. Although this analysis can be quite complicated, improving a feature to make it less trouble-prone can make a big difference to product profitability. These support-cost drivers make for an instant business case that's hard to refute in the feature prioritization process.

• An additional source of prioritization data comes from customer satisfaction and referencability metrics. These data should be stored at the opportunity level, if possible, as the satisfaction with your product or service depends on the context of the purchase (e.g., the department that bought it and their reasons for purchase). With this level of detail, you'll be able to identify the products that are most closely associated with happy customers...and the ones that seem to cause customer consternation.

Continue Reading