Trends don’t need perfect data

Trends don't need perfect data

When it comes to analyzing data trends (changes in data over time), it's important to keep in mind that the data does not have to be perfect (data never is!) in order for it to be useful.

In a conversation with a client recently, the client pointed out that their membership dashboard over several years is not 100% accurate, and therefore can't be trusted. They explained that some of the data for certain years was inaccurate (sometimes as much as only 95% accurate). And thus they didn't think the trending data was useful.

But here's the thing: the point of trending data is to see trends. Is membership growing? It is flat? Is it shrinking? Unless you're dealing with really small data sets (e.g., 50 members), a few percentage points off perfect is unlikely to affect the overall trends.

As the old saying goes, don't let perfect be the enemy of good. If you're trying to detect trends, "close enough" is very likely to get you the information you need to act.

Wes's Wednesday Wisdom Archives

Eyes wide open and affirmative decision-making

April 16, 2025

Eyes wide open and affirmative decision-making When I work with my clients on any type […]

Honoring an industry legend

April 9, 2025

Honoring an industry legend I’m taking a break from my usual data management tips to […]

Never burn a bridge

April 2, 2025

Never burn a bridge My children have reached the age where they have, or are, […]

You CAN compete with the big guys…

April 2, 2025

You CAN compete with the big guys… In my experience, associations often undersell their actual […]

Join (or start) your users group!

March 26, 2025

Join (or start) your users group! Today’s message is simple: If the AMS you’re using […]

They don’t care, until they care.

March 19, 2025

They don’t care, until they care. One of the more common questions I get from […]

Longfellow and data management

March 12, 2025

Longfellow and data management “We judge ourselves by what we feel capable of doing while […]

Do you really need all that historical data?

March 5, 2025

Do you really need all that historical data? A question I’ll often get from my […]

AI actually requires thinking

February 26, 2025

AI actually requires thinking “I don’t think AI introduces a new kind of thinking. It […]

It’s not the mistakes, but how you respond

February 19, 2025

It’s not the mistakes, but how you respond Recently a client was complaining about a bug that […]

Leave a Comment

Your email address will not be published. Required fields are marked *

Scroll to Top