Vendor consolidation has been the buzzword on every CIO agenda for two years running, but the actual cutting hasn’t always matched the rhetoric. I sat down with seven other Melbourne CIOs over the last month - mostly mid-market, $100m to $800m revenue - and asked the same question: what did you genuinely consolidate in the last 12 months, and what did you keep meaning to cut but didn’t?

The answers were more interesting than I expected.

What actually got cut

Observability was the most common win. Five of the eight had collapsed from three or more tools (typically Datadog plus an APM plus a log aggregator plus something for synthetic monitoring) down to one or two. Datadog won most of those bake-offs but two went to Grafana Cloud and one went deep on Azure Monitor because they were already a Microsoft shop end-to-end.

Average savings reported: 18-32% on the observability line. But everyone was honest that the migration cost ate the first year of savings. The real benefit shows up in year two.

The second-most-common cut was project management and collaboration sprawl. Several teams had ended up with Jira, Asana, Monday, ClickUp and Notion all running for different parts of the business. Most consolidated to either Atlassian end-to-end or Linear plus Notion. The Atlassian ecosystem in particular has become a default for Melbourne mid-market because the integration story now stretches into Loom, Trello, Confluence and Compass without the licensing pain it used to have.

Third was the AI tooling line, which is interesting because it didn’t exist as a meaningful budget line two years ago. Several CIOs had let AI experimentation run wild in 2024-25 and were now pulling it back to a managed default - usually Microsoft Copilot for the enterprise tier plus one or two specialist tools. The savings here were less about cost and more about reducing data exposure surface.

What didn’t get cut

Two patterns emerged here. First, most CIOs had a list of “should consolidate” tools that hadn’t actually moved. CRM was the standout - everyone wanted to consolidate but nobody had the appetite to actually do a Salesforce-to-HubSpot migration or vice versa. The switching cost is too high and the political cost (sales doesn’t want to move) usually kills it.

Second, niche specialist tools survived consolidation more than I expected. Things like a dedicated transcription service, a specific design tool the marketing team uses, or a particular data warehouse client - these tend to escape consolidation because they’re cheap individually and the people using them care a lot.

This matches what SmartCompany wrote in their January piece on enterprise software trends - the bigger savings now come from rationalising the middle of the stack rather than the edges.

The new entrants making consolidation harder

Worth flagging: some categories have actively expanded. Data tooling is one - most Melbourne mid-market CIOs I spoke with now run a more complex data stack than two years ago, not less. The shift to data lakehouse architectures plus the addition of vector databases for AI workloads has added tools rather than removed them.

Same story for security. Even the most aggressive consolidators kept multiple tools across SIEM, EDR, identity, secrets management, and code scanning. The CISO function has been the strongest internal lobbyist against consolidation in most shops.

Where the savings actually land

This was the part I found most useful. The CIOs who reported the cleanest savings stories had done three things consistently:

They consolidated on what they already used heavily. Not on the new shiny tool. If you’re 70% Microsoft, the consolidation is into Microsoft. If you’re AWS, you collapse into AWS-native services where possible. Going against the existing centre of gravity always costs more.

They timed consolidation to renewals. The savings in year one come from negotiating the consolidated contract harder, not from cutting licences. Multi-year commits with the consolidated vendor extracted real discounts (15-25% range was common).

They invested the savings in capability, not banked it. Every CIO who reported sustained value from consolidation had reinvested into either more headcount, a strategic project, or capability uplift. The ones who handed the savings back to finance found themselves fighting the same fight again 18 months later because new tools had crept in to fill perceived capability gaps.

A note on AI consolidation

One specific area worth calling out - there’s a real trend toward consolidating AI work with a single delivery partner rather than running multiple consultancies in parallel. Several CIOs I spoke with had moved from a “best tool for each problem” model to a single AI consulting partner that could span data, model selection, integration and ongoing support. The argument was less about cost and more about coherent architecture - having one team that knows your full AI estate.

I’m not fully sold on this yet for very large enterprises. But for mid-market it makes a lot of sense. The cognitive overhead of managing multiple AI vendors with overlapping capabilities is real and easy to underestimate.

What I’m doing next

In my own shop, the consolidation queue for the next two quarters: API gateway (we have three), identity provider (long overdue), and our analytics tooling. We expect maybe 12% in direct savings on those three lines combined, but the real prize is fewer integrations to maintain and a smaller surface for things to break at 2am.

The honest truth about vendor consolidation in 2026 is that it’s harder than the slide decks suggest and slower than the board would like. But it does work, if you pick your targets well and reinvest the upside.

Most of my peers reported they’d run another consolidation cycle in 2027. Nobody said they regretted the work.