The software world moves so fast that even Doc Brown’s DeLorean can’t keep up.

Every time you open your email – or Facebook or Snapchat or whatever new app you just asked your 14-year-old to put on your phone – you can discover a new online software system or a new option for your business-critical software.

But history tells us that not every new device or piece of software is worth integrating for your business. Established providers regularly sunset and deprecate software and apps, and startups with great ideas get bought out or fail, leaving software unsupported.

When this happens, your business is left dangling in the wind, relying on software that doesn’t work well (or at all).

But avoiding new technology isn’t the answer either. At Worthwhile, we have discussions with many companies that use legacy ERP systems that are 20 years older or more. They rely so heavily on these ERPs that change is daunting. But their employees and customers are continually frustrated by poor user experience and the inability to access things via a web portal or web interface.

New software is sometimes a necessity, and sometimes it’s a mistake. So how can your company make wise choices about when to take the leap and add a new ERP, build new custom software, or add a new product to your technology stack? Here are some companies we ask both internally and with clients.

What will we gain vs. what will it cost?

Weigh the benefits of new software against the cost of implementing it in your business. These costs can include time, efficiency, support requests, and downtime.

Sometimes, the benefits will be so large that a massive change (such as a new ERP system) make sense. At other times, a much less costly addition such as a web interface that uses an API integration to extract, transfer, and load data into your ERP will make more sense.

The same question should be asked with mobile apps. Do you need functionality that an app provides best, or will a less costly solution such as a responsive website suffice?

If the ROI makes sense, then add the new software. If not, then put off the project, whatever the size.

How much time will it take to implement?

This is a key question. Think both in terms of timeline (number of weeks or months) and in terms of training and implementation time for your team.

Our advice (and our internal approach) is usually to start simply and add features over time. This lets you recognize ROI from efficiency more quickly, and also provides real-world feedback from users about advanced features. That keeps you from wasting time and money on bells and whistles that aren’t really what your users need.

How reliable is it?

Any new technology your business adopts must be trustworthy. It should be thoroughly tested and tried before your business starts to rely on it.

So when a new app hits your radar, make sure it’s dependable. This may mean you choose not to be an early adopter, in order to preserve business continuity. That’s OK. It’s more important for your business to run smoothly than it is to have the latest app in place.

Worthwhile focuses our development on the Python language and the Django framework. Both are widely used and have vibrant communities that support and improve them.

New software can be a great thing for your business—but there’s no guarantee of it happening. So make good decisions based on ROI, efficiency, and business continuity.