Reality Bytes: Real-time analytics is a luxury most businesses can’t justify

Some ideas sound good simply because they’re technically impressive. But as with most things in IT, impressive doesn’t always mean useful.

Reality Bytes is a series where our technical experts share their opinions (we’ve called hot takes) that challenge default thinking - not to be provocative for the sake of it, but to help organisations make smarter, more intentional technology decisions.

This one often raises eyebrows.

Real-time analytics is a luxury most businesses can't justify

By Chris Beckett, Data Intelligence and Architecture Director at Inde

Personal hot take: Cereal is soup
Professional hot take: Real-time analytics is a luxury most businesses can’t justify.
Audience survey results*: 92% agree, 8% disagree 
(*Inde survey at Reality Bytes event of 36 NZ IT professionals)[

Real-time analytics sounds like something every modern business should have. Dashboards updating by the second. Metrics movinglive on screen. Data that never sleeps.

But when you step back and ask what decisions are actually being made with that data, the picture often changes.

Analysis paralysis, the challenge of real-time analytics

While real-time analytics promises faster insights and agility, studies and industry analyses show it can sometimes be counter productive or even detrimental. Research finds that overly frequent feedback can lead to worse decisions due to overreliance on short term noise. Real-time systems are expensive and complex, and they often produce more data than teams can effectively use resulting in analysis paralysis, misalignment with business goals, and lower decision quality.

So, here’s my unpopular opinions on why real-time analytics is a luxury most businesses can’t justify:

Not every decision needs to be made right now

Your CEO probably doesn’t need to know your website conversion rate every five seconds.
Your supply chain manager doesn’t need an alert the instant inventory moves by one unit.
Your online retail manager doesn’t need a live counter ticking over as orders hit the warehouse.

What they usually need is accurate, timely insight, daily, weekly, or monthly that helps them make the right decision, not the fastest one.

My advice is often look at those decisions that you need to make, understand how often you need to review the data and then work backwards from there.

Real-time is a tax and it’s an expensive one

I often describe real-time analytics as a tax. Not because it’s bad, but because it’s costly and frequently misunderstood.

The moment you move from batch or scheduled reporting into real-time, complexity increases sharply. Systems need to run continuously. Data pipelines must handle constant ingestion. Reports change while people are looking at them. Costs rise especially in cloud environments where you pay for compute and processing around the clock.

Batch ingestion, by contrast, is simple, predictable, and efficient. It runs on a schedule, delivers stable outputs, and is easy to maintain. For most business use cases, it does the job perfectly well.

Real-time platforms don’t just cost more to build they cost more to operate, monitor, and explain.

So when does real-time analytics really matter?

This isn’t an argument against real-time altogether. There are absolutely scenarios where instant data is essential.

A good example is emergency response. When river levels rise rapidly, like we saw recently with the Waimakariri River in Christchurch, real-time data matters. You need to know now, because action needs to happen now.  That’s the key distinction.

If the data triggers an immediate response, real-time is justified. If it doesn’t, it probably isn’t.
By contrast, knowing how many people walked through the door at an event, second by second, isn’t actionable in the moment. But reviewing attendance patterns later to assess how the event performed? That’s incredibly valuable.
Same data. Different timing. Much lower cost.

Start with the decision, not the dashboard

The sweet spot is working backwards from the decision you’re trying to make.

Ask:
•    What decision will this data inform?
•    How often is that decision made?
•    What’s the real cost of the data being reported in a day, a week or a month?

If your leadership team meets once a month, make sure the management pack is ready for that meeting. If an account manager reviews performance before a client check-in, make sure the data is available then.  If the decision isn’t happening right now, the data doesn’t need to be real-time.

Right-time beats real-time

The most effective data platforms aren’t the fastest they’re the most appropriate.

By focusing on right-time analytics, organisations can:
•    reduce cost and platform complexity
•    improve trust in reporting
•    align data delivery with real decision-making
•    invest effort where it delivers genuine business value

The real challenge is resisting the hype and choosing the level of freshness that actually supports how your organisation works.

Before your next real-time analytics project, ask one final question:
What would it really cost if this data arrived later and what could we do with the time and money we save?  That’s where smarter data strategy starts.