old adding machine
Change is evolutionary, and some legacy solutions are still the right solutions for a given activity. PHOTO: Andrew Branch

Businesses have their favorite analytic software — tools entire processes and operations have relied upon for years. 

But change is inevitable. Software does not last forever, especially as new data management options arrive in the marketplace.

So the ultimate question is: When should professionals replace existing analytic software?

That question is starting to arise more frequently as managers at all levels of an organization seek more people with technical acumen and training on the latest tools. 

Thus tools for a given set of skills and operations may hold back managers who want to effectively move organizations to the next level of analytics.

But not every tool is immediately outdated. Change is evolutionary, and some legacy solutions are still the right solutions for a given activity.  

So-called industry experts, seeking to entice marketers with a shiny new bauble, may post about a solutions’ imminent demise. Instead that tool flourishes with new possibilities.

4 Questions to Consider

So when does a manager know when it's time to let go? Here are some signs to look out for.

Is the role of current tools evolving?

Some tools are not exactly outdated, although their purpose has altered over time due to new design features that allow tasks that could not be done in other software. 

Take Microsoft Excel. It has been widely adopted in every business. Despite its widespread use in business, some experts claim its feature limitations make its use questionable for big data strategy.

For example, Excel limits the number of rows in a spreadsheet to about one million, less than what tables created in R programming can import from many data sources.   

But Excel can aid in other tactical ways.  Plugins allow users to import data — Supermetrics, for example, extracts Google Analytics data. The data volume is small enough to permit audits and can display data anomalies more freely than what the standard analytics interface can reveal.

In short, marketers have to weigh whether software itself is outdated or just the utility of some of its features.

Are tools underused?

People who deeply understand the strategic objectives of the business will identify the technical needs that support those objectives. 

Sharing that understanding leads to a consensus on what skills are needed, but it can also raise a consensus on how tool features should be used.

In his book Competing on Analytics, Thomas Davenport wrote: “Without a distinctive capability (what you do to set your business apart), it becomes impossible to compete and distinguish what data is important.” Given the rise of analytic solutions in the marketplace, the same can be said for tools.

If there are idiosyncrasies among employees regarding a given tool, team leaders must investigate the reasons behind that and be proactive in implementing changes in use behavior. Determining better use of solutions is a core challenge, because the solution — training — is straightforward.

Is assessing data quality getting easier or more complicated?

Modeling data and understanding capability, however, is not straightforward.  A model is planned with a consideration to what data relationships should look.

The best “forward-thinking” tools should highlight when missing data can impact the accuracy of models created with the data. For example, some programming languages that are used extensively with datasets, such as R and Python, have file validation plugins designed to automate validation tasks. 

That automation can save time in parsing data, and can help reveal if different tools or data science skills are ultimately needed for an overall data strategy. 

When analysts clean data, they should learn in the process what effort needs to be simplified to gain better data quality. 

Can your team establish consistent communication around the tool’s results?

The ability to express ideas behind the data is not always perfect. This is especially true with analysts and teams who work remote from each other. Large gaps in communication with remote workers can occur.

Determine if the analytics tool features encourage better communication of project hand-offs. Project hand-offs can muddy the details behind the “why” of tasks  — topics that are often communicated among team members.

Talking to teams can reveal if current tools encourage robust conversation that helps cross-functional teams understand critical concepts behind decisions.

Seek a 'Coach' for Implementation

Even the most talented, successful and hardworking athletes have coaches to give an outsider’s viewpoint and raise potential blind spots in executing a given plan.  

Organizations evaluating their performance within different departments can take a similar approach.

Consulting firms with technological expertise in business intelligence solutions can play the role of “tech coach” for an organization.

Firms can create a map of tools, teams using tools and concerns to assess how the organization can better integrate solutions that support overall objectives and help employees be more effective.

Analytics is not riding into the sunset anytime soon, but software for that purpose can go away — be it by poor development support or even by mergers among the very providers themselves. 

Organizations can avoid complication from the wane of software by periodically evaluating how their tactics, tasks, and software usage evolve over time.