Relative scoring does have merits, especially when comparing to other products available at the time scored. However I'm afraid a product reviewed highly 3 years ago (with low competition) will look better than a new product that is scored lower against a much higher relative standard.

There are a couple approaches that might mitigate stale relative scores:

1. Score against absolute benchmark, then translate to current relative index. Whenever the relative index is adjusted to higher standard the old scores will automatically decrease compared to new products.

2. Develop relative improvements table that can be used to adjust scores in each area.(Tricky as new metrics are added that old product review might not have information about.) For example, perhaps one metric index has improved 20% in the industry over last 2 years. All reviews 2 years old would take a 20% decrease in their original score for that one metric. The table of relative adjustments needs to be published including adjustment amount, reason for each index adjustments, and official adjustment date. A link to this relative adjustments table needs to be added to any review over 6 months old (or provided in ALL reviews with understanding that new reviews are unaffected).

3. Mark all products reviewed over 2 years old with "Stale review" warning stating readers should not directly compare with current scores. Then re-review highest scoring and most popular (by sales volume) products every 2 years to update their score to current standards (and remove "stale review" warning flag).

What is the SmartGridNews.com plan for responsibly managing relative review scores over time?