Who holds the leverage when you choose a tool

The thread today is leverage: who holds it in the relationship between a business and the software it runs on. Three reports show it from different angles — a spending forecast where price rises carry much of the growth, an outage that ended without a credit, and a one-person company that reached real scale in under a year.

When the price rise does the work

Software budgets move in two ways. You add tools and seats, or the tools you already have cost more at renewal. The second kind is quieter. Nobody signs a purchase order for it. It arrives as a line on a renewal letter, and most teams pay it because switching is harder than absorbing the increase.

Gartner's latest forecast puts numbers on this. The firm expects worldwide IT spending to cross a threshold it has not reached before [1]. Part of that growth is new capacity — data centres, models, AI application platforms. But a meaningful share is simply existing customers paying more for the same thing. When a market grows and price carries part of that growth, your renewal is not an afterthought. It is the growth engine for the vendor on the other side.

For a business choosing tools, the practical lesson is to budget for the increase before it lands. Assume the number on next year's renewal is higher than this year's, and decide now what increase would make you look elsewhere. That threshold is easier to set when you are calm than when the invoice is in front of you. It also helps to know which tools you could actually leave. A tool with your data locked inside it and your workflows built around it has pricing power over you, whatever the headline rate says. We wrote more about reading the figures that should move a decision in numbers that change a decision.

An outage, and the silence after it

The second report is a first-hand account. A critical application went down for close to two full days [2]. Work stopped. The team built workarounds in spreadsheets, and someone asked for an update every few hours [2]. What the writer remembers is not the downtime itself — outages happen to everyone — but that no credit was ever offered.

This is worth teaching in plain terms, because the mechanism is often misunderstood. A service-level agreement is the vendor's promise about uptime. A service credit is what you are owed when they miss it: usually a small refund, proportional to the downtime, written into the contract. The credit is rarely large enough to cover the real cost of lost work. Its value is as a signal. A vendor who writes credits into the contract and pays them without being asked is telling you they expect to be held to their promise. A vendor who offers nothing, even after two days, is telling you something too.

So before you make a tool critical to your operation, read what it owes you when it fails. Find the uptime number. Find the credit terms. Find out whether the credit is automatic or whether you have to claim it. If none of that is written down, treat the tool as best-effort, and plan a manual fallback for the day it is unavailable. The spreadsheet the team reached for in this story is the honest version of that fallback — a sign the work had outgrown a safety net nobody had planned. We looked at the point where that improvised sheet stops being enough in when a spreadsheet stops being enough.

There is a second lesson in the silence after the outage. How a vendor behaves on its worst day is part of what you are buying. It rarely shows up in a feature comparison, but it shows up the first time something breaks.

One founder, two hundred million

The third report is about speed. The piece opens by saying that every few years a company comes along that rewrites what founders think is possible [3]. Its subject is Base44: by the account's own figures, a company started by a single founder, with no outside funding, that drew eighty million dollars from Wix within six months and reached two hundred million dollars of annual recurring revenue. The founder, Maor Shlomo, was thirty-one and coming off reserve duty when he began in January 2025.

The detail that matters for a buyer is the timeline. A product can now go from one person to serious scale in roughly a year. Ten years ago that path took a team, a funding round and several years. The reason to care is not the valuation. It is that the market you are shopping in now contains far more young, fast-moving vendors than it used to, and they will reach you sooner.

This is a genuine trade-off rather than a simple good-news story. A vendor at this speed can ship features faster than an established one and often prices to win. A vendor this young also has a shorter track record — fewer renewals behind it, fewer outages survived, less history of how it treats customers when something goes wrong. Which brings the day full circle. The reliability question from the second report is exactly the question a fast-growing new tool cannot yet answer, because it has not been around long enough to answer it.

Reading the three together

Each report is about the same thing from a different side. The forecast says prices on existing tools will keep drifting up, so the cost of staying is rising [1]. The outage says the cost of a tool failing is rarely covered by what the vendor owes you, so you carry most of that risk yourself [2]. The new entrant says there are more alternatives arriving faster than before, but the newest ones have the least history behind them [3].

Put together, they describe a buyer's job. Know what you could leave and at what price. Know what each critical tool owes you when it breaks, and have a fallback ready. And weigh a new vendor's speed against the years of evidence it has not yet had the chance to build. None of this requires a bigger budget. It requires reading the contract and the trend before the renewal, not after. If you want a fuller checklist for that judgement, we set one out in choose software worth using.