Software ate the world. Now it’s eating software companies.

Do you want excellence, or convenience?

Cloud computing allowed a generation of enterprise software companies to flourish, providing them the freedom to tinker with laser-focused ideas that could streamline stodgy business workflows within a browser or an app. That has led to a proliferation of different enterprise software vendors offering high-performance but niche products: According to data from Okta , which makes identity management software, its average corporate customer is using 88 separate apps across their business.

One of the main things that made this shift possible was the decentralization of technology purchasing ushered in by the cloud. That allowed the leaders of different business units — from marketing to HR, purchasing to legal — to pick and choose the technology tools best suited to their team's needs. Those leaders were fed up with the top-down approach that dominated the early years of information technology, when a senior executive would pick a suite of software that everyone was forced to use regardless of whether it made sense for a given business unit.

Fast forward a decade, however, and some of the most in-demand enterprise software businesses help their customers keep track of all the money they're spending on different software tools across different vendors. And depending on the pace of the economic recovery from the effort to fight the pandemic, the next few years could be lean ones inside corporations around the world, potentially forcing tech buyers to pick and choose just which of those many software products they really need.

That presents an opportunity ripe for the massive enterprise software players of the world — Microsoft, Salesforce, Workday and ServiceNow, to name a few — to continue acquiring smaller enterprise software companies while building an integrated platform of all the tools needed to run a modern business. Just this week, we've seen ServiceNow buy Element AI , and, of course, Salesforce's massive $28 billion bet on Slack .

And so there looks set to become a tension: one of balancing cost savings from buying good-enough software from the giants against the inexorable move to department-level IT purchasing that prizes functionality above all else. In other words, technology leaders will have some decisions to make about how they procure software over the next several years.

"My focus is on ensuring my employees are successful and productive wherever they are," said Alvina Antar, CIO of Okta. "Obviously, I am also committed to driving efficiencies [but] within the lens of ensuring my employees are empowered to be successful."

Good enough

The past decade was a golden age for enterprise software delivered as a service, either through a browser or a mobile app. Salesforce and others had proven the concept could work. Then the maturation of cloud infrastructure services just as companies were recovering from the Great Recession allowed enterprise entrepreneurs to very easily start successful companies.

Those founders usually focused on building one type of enterprise service — file storage, payroll, help desk management — and building it well, moving faster than more-established enterprise software companies such as IBM, Microsoft and Oracle to meet a new generation of workers where they wanted to be. More and more, those purchasing decisions were being made by the heads of vertical departments like marketing or finance who had a far better eye for quality and utility than a senior executive who was unfamiliar with the day-to-day needs of their group.

"People are by default going to SaaS in general. Nobody's saying, 'Hey, let's spin up a bunch of developers, or hire IBM to build this for us,'" said Jerry Chen, a partner at Greylock who has backed cloud upstarts such as Docker, Blend and Notable.

Many of those entrepreneurs now run large public companies and find themselves at the confluence of new trends.

Subscription fatigue is setting in among some software buyers, where the need for cost efficiency suddenly became a lot more important this year. Even Microsoft, which corrected its course after Satya Nadella took over and is riding high on the surge of interest in cloud services, asked employees in its security division last year to reduce the number of different SaaS tool vendors they used to reduce management burden.

At the same time, early SaaS companies that were built around a single product or service are branching out into new areas in search of growth. Salesforce paved the way for this market, and it has grown far beyond its sales-productivity roots by acquiring dozens of companies over the last decade, most recently adding Slack and shelling out $15.7 billion last year to acquire business intelligence specialist Tableau.

That activity gives the enterprise software buyer something new to think about. For many years, the focused SaaS companies — here's the obligatory "best of breed" mention — were simply much better than the "good enough" alternatives from the largest enterprise players.

"The reason why we invest in best of breed is because we're betting on innovation," Antar said. "We're betting that the person that's living and breathing the capability that we're investing in will innovate much faster than the consolidated solution, because it's not [as much of] a focus."

But a flurry of acquisitions over the last several years — it's not just Salesforce — has brought some very talented enterprise product executives into the larger players, in hopes of helping them stay current and encouraging their customers to buy new, different services from a vendor they trust. Google Cloud's acquisition of Looker, SAP's deal for Qualtrics, and Twilio's acquisitions of SendGrid and Segment are all part of the same movement.

Learning from the best

There will always be enterprise software experts with the itch to start their own business, ensuring this cycle will continue for years to come. But it's certainly now the case that the product gap between startups and larger players — like Microsoft, Google Cloud and Workday — has narrowed.

The largest software companies have the most data, and they are getting better every year at extracting insights from that data. As they build a platform of enterprise software, they can design new products based on those insights as well as the best level of interoperability between those products.

Take Salesforce and Tableau: Salesforce already offered a business-intelligence tool called Einstein Analytics when it announced the deal for Tableau, but has since redesigned that product around Tableau's vision. Einstein Analytics, however, was designed to work in concert with Salesforce data stored in its Sales Cloud or Support Cloud products, and blending the two will theoretically give Salesforce customers a better analytics product that still works closely with all their existing data.

Yet there's work to be done. If the big companies want to disrupt a whole generation of thinking about enterprise software buying, they will need to get far better about actually integrating data between the startups they acquire and their most popular services, such as Sales Cloud or Microsoft 365 or Twilio Flex.

As a customer of both Salesforce and Tableau prior to its acquisition, Antar is waiting to see how that promised integration plays out.

"The vision would be for us to be able to take advantage of the fact that a lot of our source of truth data sits within these clouds, that it allows us to be able to then understand and drive the quality of data, and [improve the] accessibility of the data, much more seamlessly as one platform," she said. Without that integration — which is harder to do than it sounds during the analyst call announcing a new acquisition — enterprise software buyers like Antar will continue to be cautious about acquisitive platform companies.