When individual contributors (ICs) and decision makers (DMs) describe “the problem,” they are often talking about different things. ICs talk about the incidents they touch. DMs talk about the risks they own. Our State of Development 2025 survey puts those two views on the same page so teams can plan with the same facts and stop talking past each other.
TL;DR: ICs report daily friction. DMs see business exposure. Both groups point to the same fix: more reliability, less manual effort, and decisions that reflect how tool choices actually get made.
What ICs feel, and what DMs fear#
ICs live in the failure path. The most common pain points are latency and performance (31%), complex workflows that feel fragile (31%), debuggability and observability gaps (28%), and manual failure recovery that still shows up in too many places (19%). That list reads like an on‑call rotation.
DMs zoom out to the risk register. Security ranks first (36%), followed by technology complexity (33%), integration friction (32%), and the double bind of cost management and scaling (both 30%). Reliability (28%) remains near the top, which tracks with what customers notice and what boards ask about.
Both are right. They are the same story at two time horizons. IC symptoms today become DM risks tomorrow if they aren’t addressed.
Operations are still noisy#
Only one in four teams describe work operations as smooth. The rest point to high operational overhead (35%) and complex failure recovery (34%) as routine. Long‑running work continues to be a sticking point, landing around the 30–35% mark depending on role. The pattern is all too familiar: teams can start work, but finishing cleanly without human babysitting is hard.
Satisfaction numbers echo that gap. Orchestration gets 84% satisfaction while automation sits at 77%. Orchestration helps you kick things off. Automation keeps them alive, observable, and recoverable when reality intervenes. Most organizations haven’t closed that second mile.
Who actually picks the tools#
Power shifts with company size. In smaller orgs, developers often drive tooling decisions. Overall, developers show 50% influence in selection. In enterprises, gravity moves upward: CIOs and IT managers lead at 53%, and developer influence drops to 35%.
That shift explains a lot of stalled rollouts. If you’re pitching a tool the same way at 80 employees and at 8,000, you will miss the mark. The buyer changes, the evidence they need changes, and the story must change with it.
What each side optimizes for#
When you ask what matters in tool selection, the lists rhyme but differ in emphasis. DMs weight security (67%), reliability (63%), cost efficiency (55%), and ease of integration (49%). This is portfolio thinking: minimize exposure, keep systems steady, prove the spend.
ICs put security (57%) in the top three as well, then lean to performance (51%), reliability (49%), flexibility and customization (49%), and integration with other tools (47%). That’s day‑two thinking: can we debug it, tune it, and evolve it without brittle glue?
Both lists point at the same destination. The fastest way there is to translate proposals into both languages. “Durable retries reduce MTTR” speaks to the IC. “Fewer incidents and lower overtime spend” speaks to the DM. It’s the same improvement framed for the accountability it serves.
What failure actually costs#
Decision makers draw a direct line from incidents to business outcomes. Customer churn or dissatisfaction leads at 48%, followed by increased operational costs (47%), revenue loss (45%), and brand damage (45%). Only 5% say failures would have “no major impact.” Reliability is not a tooling opinion. It’s a revenue, retention, and reputation issue with a clear price tag.
This is where the IC list and the DM list meet. Manual recovery and opaque flow logic turn into missed SLAs and support spikes. Better execution models and shared visibility turn into fewer tickets and calmer quarters.
The near‑term North Star#
Across roles, the next 12–24 months cluster around three goals: enhance reliability and security compliance (36%), increase automation and workflow efficiency (33%), and reduce operational costs and technical debt (30%). That is a workable roadmap you can bring into planning. It matches what ICs need to ship and what DMs need to report.
Where Temporal fits#
The report shows meaningful adoption of Durable Execution already: 49% overall and 60% in large companies, with only 8% unfamiliar. That maps cleanly to what both groups want. ICs get a programming model that completes work even when systems restart. Retries, backoff, and state persistence come standard, so workflows become easier to debug and safer to change. DMs get higher reliability without a growth in operator headcount. Incidents fall, recovery times improve, and the cost curve bends the right way.
Temporal was built around that model. You write straightforward code for business logic and let the platform provide the survivability. It’s a pragmatic way to move from “we can start it” to “we can finish it every time,” which is the difference customers feel.
Get the full picture#
This post is the high‑level read on where ICs and DMs agree, where they diverge, and how to use the split to make better decisions. The full State of Development 2025 report includes the segment cuts by role, company size, and region, plus the charts behind each number here. If you’re planning next quarter’s reliability work or making a tooling case, it belongs in that meeting.