Has your data lake changed from pristine, clear waters into a murky, muddy swamp? You’re not alone. At Gartner’s recent Data & Analytics Summit Americas conference, analyst Donald Feinberg highlighted that many companies now find themselves in this situation, struggling to save their data lakes. Luckily, as he details in this recent article from Datanami, he has a life preserver to throw them.

Feinberg offered companies a few tips for cleaning up their data lakes, among them:

Avoid Huge Implementation Projects

Many companies get into trouble with data lakes early on by committing to a long-term project and putting everything into one big data lake. A better approach is to start small, ideally implementing a data lake for a single business unit. Of course, it’s important to remember that these small data lakes are a component of the larger picture and that eventually you will likely require multiple data lakes. But Feinberg believes that keeping them small and separate, at least to start, has a better likelihood of success than attempting a large-scale data lake project.

Don’t Equate Having a Data Lake with Having a Data and Analytics Strategy

A data lake may certainly be a component of a larger data and analytics strategy, but it is not a strategy in and of itself. According to Feinberg, many companies will often fall into the trap of conflating the two, which leads to confusion and can curtail a data lake project’s success.

Focus on Business Value

Another important consideration is to ensure that the prospective business value of the data lake project is always top of mind. After all, if you can’t identify what actual business value the data lake will bring to the organization, there’s a good chance you’re fishing in the wrong lake.

There’s No Such Thing as an Infinite Data Lake

Because of data lakes’ flexible nature, companies tend to store a variety of data comprising different formats and sources. Feinberg cautions against this “infinite data lake” mentality, as it often leads to significant governance headaches.

Address Skill Deficiencies

Feinberg believes that one of the most important things organizations can do to avoid data lake failures is to close the talent gap. Finding and honing the right skills will enable organizations to pull themselves out of the murky waters and identify the tools and technologies that will ensure that, once cleaned, their data lakes never become swampy again.

For more on the above and some additional advice from Feinberg, read the full Datanami article.