2026 State of Analytics Engineering Report

Analytics engineering, accelerated

Analytics engineering has entered a new phase of maturity. Artificial intelligence is no longer experimental inside data teams. Instead, it's funded, embedded, and actively reshaping analytics engineering workflows, influencing how code is written, how insights are generated, and how teams invest in analytics and data infrastructure. What was once exploratory is now operational.

In 2026, the field is defined by AI-driven acceleration and the pressure it creates. At the same time, the core challenges that have long defined analytics engineering—data quality, ownership clarity, governance discipline—remain largely unchanged.

AI is expanding what analytics teams can build and deliver. But the reliability of those outputs depends on the systems that govern them: validation, clear ownership, and strong data controls.

The 2026 State of Analytics Engineering Report highlights a defining dynamic: AI is scaling analytics output faster than the trust and governance mechanisms designed to support it.

The central question for 2026 is whether data teams can meet growing demands for speed and productivity without compromising data quality and trust.

Here’s what we found

AI-assisted coding is now embedded in daily workflows.

72% of respondents now prioritize AI-assisted coding in their development process.

Trust and speed are emerging as the top performance priorities

The importance placed on trust in data and data teams rose from 66% to 83% year over year, while the importance placed on speed increased from 50% to 71%. At the same time, fewer teams cite lack of trust in data as a top operational challenge (33% to 24%), suggesting that trust is becoming a strategic expectation rather than a recurring issue.

Governance concerns remain elevated.

71% are concerned about hallucinated or incorrect data reaching stakeholders, and 41% cite ambiguous data ownership as an ongoing challenge.

Data infrastructure costs are outpacing budget growth.

57% report increased warehouse and compute spend, compared to 36% reporting increased team budgets.

2025 → 2026: What changed

The shift from 2025 to 2026 is less about expansion and more about consolidation. What accelerated last year is now becoming embedded in everyday analytics workflows.

Three patterns define the transition:

1. Acceleration → Integration


AI-assisted coding has moved from rapid adoption to embedded practice.
What surged in 2025 is stabilizing as a baseline workflow in 2026.

2. Enablement → Reliability


Trust has become a clear strategic priority, even as structural challenges like ambiguous ownership remain largely unchanged. The focus is shifting from expanding access to reinforcing reliability.

3. Speed vs. cost → Speed vs. trust


Speed increased significantly year-over-year, while cost priorities saw only marginal growth. Performance expectations are no longer defined primarily by efficiency; they're increasingly defined by the ability to maintain trust and reliability while moving faster.

2025 revealed potential. 2026 requires discipline.

The 2026 State of Analytics Engineering Report explores how data teams are operating under sustained acceleration. It captures where priorities are shifting, where constraints persist, and what those signals reveal about the next phase of analytics engineering.

Key Insights

AI adoption is reshaping analytics engineering

Defining AI in this report

To ensure clarity, this report distinguishes between two forms of AI use:

AI assisted coding

LLMs used by practitioners to draft or refactor analytics code (SQL, Python, YAML), tests, and documentation.

AI-generated insights

Stakeholder-facing outputs produced from natural-language prompts.

When “AI” is used in section headers, it refers broadly to both categories. In analytical sections, we specify which layer is being discussed.

AI is embedded in analytics workflows

AI is no longer exploratory inside analytics teams; it’s embedded in daily workflows.

A clear majority of respondents (72%) prioritize AI-assisted coding within their development process. Among leaders, more than 77% emphasize AI for productivity gains.

LLM usage is now embedded in analytics development workflows, primarily reducing cycle time from draft to production.

AI-assisted coding has become a strategic focus, though investment remains uneven across the pipeline.

While 72% prioritize AI-assisted coding, only 24% prioritize AI-assisted pipeline management, including testing, observability, and quality controls. The signal is clear: teams are leveraging AI to accelerate creation, but not to reinforce governance at the same pace.

This isn’t a distinction between coding and pipeline work. AI-assisted coding contributes direction to pipeline development.

The contrast is between acceleration and stabilization:

• Acceleration: generating models, transformations, and documentation faster.
• Stabilization: reinforcing validation, testing, observability, and governance controls within the data pipeline

AI investment is currently weighted more heavily toward acceleration.

Governance pressure is rising

As AI adoption expands across analytics workflows, governance readiness is becoming a critical concern. Several signals in this section point to a growing gap between acceleration and the systems designed to validate and govern AI-assisted outputs.

As AI expands what analytics teams can produce, concern is rising alongside adoption. This raises a central governance question for 2026: whether validation, testing, and oversight mechanisms are scaling at the same pace as AI-driven output.

Risk awareness remains elevated

Adoption has increased, and so has awareness of the risks that accompany it.

A significant majority of respondents (71%) are concerned about hallucinated or incorrect data reaching stakeholders. However, the intensity of concern varies across roles.

Practitioners report higher levels of concern around implementation risks. They show a 7-percentage-point higher level of concern than leaders about exposing sensitive data to LLMs.

They also report elevated concern around the quality implications of AI-assisted development. By contrast, leaders are more likely to frame challenges in terms of compliance, documentation, and governance preparedness.

The pattern suggests proximity matters. While practitioners experience execution risk directly, leaders experience governance responsibility structurally.

AI adoption is accelerating across two fronts: engineering throughput driven by AI-assisted coding, and stakeholder-facing outputs through AI-generated insights. However, investment in validation, testing, and governance mechanisms isn’t scaling at the same rate.

The imbalance is clear. Key indicators of AI acceleration, such as prioritization of AI-assisted coding, sit in the 70-80% range, with trust carrying similar strategic-weight (83%). Yet governance investment and foundational data constraints remain uneven.

Output is scaling faster than stabilization.

What this means

LLM usage across analytics workflows has moved from experimentation to implementation. Its primary impact today is operational—helping teams ship code and analysis faster by reducing cycle time from draft to production.

AI is scaling across two dimensions: increasing engineering throughput and expanding stakeholder-facing outputs. However, investment in validation, testing, and governance mechanisms isn’t scaling at the same rate. As delivery speeds increase and AI-generated insights reach broader audiences, the controls designed to safeguard that output are maturing more slowly.

Acceleration without stabilization compounds risk.

Join us live on April 29 or 30 for the 2026 State of Analytics engineering virtual event to go deeper on these AI findings with Katie Bauer (Hex), Jay Sobel (Ramp), and Jason Ganz (dbt Labs), as they share how leading teams are operationalizing AI without compromising reliability. You’ll walk away with practical ways to balance speed with trust, apply AI beyond just coding, and build the controls needed to scale AI-driven analytics—with the chance to ask your questions live and win a few fun giveaways.

Trust and speed are emerging as the top performance priorities

Performance expectations for analytics teams are shifting. In 2026, trust and speed are the most emphasized priorities, even as cost pressures remain present.

The share of respondents who say increasing trust in data and data teams is important rose from 66% in 2025 to 83% in 2026. The importance placed on speed also climbed sharply, from 50% to 71%. Cost reduction, by comparison, saw only modest movement, edging up from 48% to 53%.

The importance placed on trust and speed is increasing more rapidly than any other objective measured, signaling a shift in how analytics performance is defined: not just output, but reliable output delivered quickly.

The combined concentration of “Important” and “Very Important” responses in 2026 reinforces this shift. Trust now carries the highest overall prioritization across objectives, and its year-over-year increase is now more pronounced than speed or cost.

Analytics teams have historically been evaluated heavily on output volume and efficiency. In 2026, that’s no longer sufficient. As AI-assisted coding and AI-generated insights increase the speed and scale of delivery, the expectation isn’t simply faster insights; it’s faster insights that stakeholders can depend on.

Cost discipline remains relevant, but it’s no longer the primary performance priority. Velocity alone is insufficient; organizations expect faster delivery paired with greater confidence in the results. In this context, maturity is defined by reliability under acceleration—the ability to move quickly while maintaining accuracy, consistency, and trust.

What this means

Speed and cost remain important, but trust now anchors both. As AI-generated insights increasingly reach stakeholders, reliability becomes part of how analytics performance is judged. Organizations are no longer optimizing primarily for output or efficiency; they’re expected to deliver insights that stakeholders can depend on at increasing speed.

Integration has improved; trust constraints persist

Not all challenges in analytics engineering are moving in the same direction. Some operational friction is easing, but structural constraints remain.

The share of respondents citing “integrating data from various sources” as a top challenge declined from 35% in 2025 to 27% in 2026. Technical integration, once a defining pain point, appears to be becoming more manageable for many teams.

Meanwhile, other signals have remained remarkably consistent.

Ambiguous data ownership remains a challenge for 41% of respondents, effectively unchanged year over year. Poor data quality continues to be the most frequently reported obstacle across organizations.

Data literacy among stakeholders remains a barrier for 36% of respondents, down slightly from 39% last year, reinforcing that trust gaps extend beyond technical systems into organizational dynamics.

Integration challenges are declining, but ownership, quality, and literacy constraints persist, shifting the bottleneck from infrastructure to accountability.

Perceptions of these challenges also vary by role.

While most challenge perceptions are broadly aligned across roles, differences emerge in select areas, reflecting variations in day-to-day responsibility and oversight.

Notably, fewer teams now cite “lack of trust in data from stakeholders” as a primary challenge (33% to 24% YoY), even as trust rises sharply as a strategic priority.

What this means

Improvements in technical integration haven’t eliminated deeper trust constraints. While fewer teams describe trust as an acute operational issue, more now frame it as a strategic expectation, signaling a shift from isolated friction to structural accountability.

The work of maintaining quality, clarifying ownership, and aligning stakeholders hasn’t disappeared; it’s become foundational. As acceleration increases and trust expectations rise, unresolved quality, ownership, and literacy gaps carry greater consequence.

Analytics engineering is evolving from an enablement layer to a control layer, responsible for ensuring that intelligent systems scale reliably, not just rapidly.

Budgets are growing—data infrastructure costs are growing faster

Investment in analytics engineering continues, but growth is uneven.

In 2026, 36% of respondents report increased team budgets, while 28% report no change, 14% report decreases, and 19% are unsure of their team’s budget status.

At the same time, spending on analytics and AI infrastructure—such as data warehouses, compute resources, and data management software—is rising more broadly.

A majority of respondents (57%) report increased warehouse and compute spend, while only 13% report decreases.

Budget growth and infrastructure growth aren’t moving in lockstep. More teams report increased warehouse and compute spend than report budget expansion.

The gap becomes more visible when comparing the full distribution of responses across the past 12 months.

While both budgets and infrastructure spend have increased for many teams, warehouse and compute spend is more concentrated in higher growth bands, reinforcing the uneven pace of expansion.

When infrastructure costs rise faster than overall budgets, teams operate under sustained financial pressure.

As workload intensity and data consumption expand, compute demand grows alongside them. Cost optimization, therefore, isn’t a signal of contraction; it’s a necessary response to sustained scale.

Future investment priorities

Leaders are more likely to prioritize increased investment in data tooling over the next 12 months, reinforcing their focus on long-term capability building alongside budget expansion.

This difference reflects role responsibilities. Leaders typically focus on longer-term investment planning and platform capability, while practitioners are more closely tied to day-to-day execution and operational constraints.

What this means

While more than a third of teams report budget growth, a significantly larger share report rising warehouse and compute spend. As AI-driven workloads expand, infrastructure demands are absorbing much of that investment. This puts increased financial pressure on analytics teams and makes investment decisions more consequential.

To stay in control, teams are prioritizing incremental processing, query optimization, and cost visibility. Efficiency is no longer optional; it’s essential to sustaining growth at scale.

Cost optimization remains a priority as teams work to sustain scale under increasing workload intensity. Financial discipline in 2026 is about ensuring growth remains sustainable, not simply cutting back.

Architecture interest exists; adoption remains limited

Teams are actively evaluating modern architectural patterns, though broad production adoption remains limited.

Interest in open table formats and multi-engine interoperability is now visible as a strategic consideration in complex data environments. Apache Iceberg, included in this year’s survey as one signal of that shift, shows early but meaningful engagement.

In 2026, only 9% report using Iceberg in production, while 6% are planning adoption and 12% are in proof-of-concept. In total, 27% report some level of engagement. Meanwhile, 68% report no current plans.

Among teams evaluating or adopting Iceberg, several motivations are driving interest. The most frequently cited driver is multi-engine compatibility (22%), followed by flexibility and performance considerations.

At the same time, organizations evaluating Iceberg report several concerns that may slow adoption. The most frequently cited barriers are knowledge gaps (27%) and unclear use cases (27%), suggesting that teams are navigating implementation complexity rather than resisting the concept itself.

Interoperability and multi-engine coordination now carry greater strategic visibility, even as most teams remain in evaluation or early adoption phases as they assess maturity, fit, and long-term implications. Architecture is entering the strategic conversation in 2026, though it hasn’t yet reshaped day-to-day operational behavior.

What this means

While more than a quarter of teams are exploring Iceberg, fewer than one in 10 have moved it into production. Interest in open table formats and multi-engine interoperability is present, but most organizations remain in evaluation or experimentation phases rather than full architectural transition.

The data suggests that Iceberg is being considered as a strategic, forward-looking investment rather than an urgent operational shift. Engagement reflects long-term planning and future-proofing behavior, while limited production adoption indicates that teams are weighing maturity, tooling readiness, and internal capability before committing at scale.

In the context of 2026 priorities, architecture decisions function as a secondary maturity signal—important, but not yet a defining driver of day-to-day operational change.

What’s next: scaling trust in an era of acceleration

The 2026 survey reflects analytics engineering operating under sustained acceleration. AI is expanding output faster than governance maturity is evolving to support it.

Today’s environment is defined by simultaneous momentum and constraint:

• AI-assisted coding is embedded in development workflows
• AI-generated insights are reaching stakeholders at increasing speed and scale
• Infrastructure demands are rising
• Trust expectations are intensifying
• Data quality, ownership ambiguity, and governance gaps persist

As a result, analytics engineering now carries the responsibility of preventing AI-generated insights from amplifying existing trust gaps. The defining production advantage in this environment will come from reliability, not volume; at scale, under scrutiny, and with clear governance.

The next phase of AI integration is likely to move from generation to execution, with systems coordinating workflows, validating outputs as they act, and operating across tools with reduced human intervention. If quality controls, review practices, and governance mechanisms don’t mature alongside this shift, autonomy will increase complexity rather than reduce it. Therefore, discipline becomes the prerequisite for autonomy.

Analytics engineering is increasingly defined by accountability—by the ability to scale trust alongside output. As new capabilities expand what teams can build, trust often determines how far that expansion can scale. Organizations that succeed in the next phase will treat trust as infrastructure, embedding it into governance, data quality, and operating rhythms.

In 2026, the differentiator is discipline.

Methodology

dbt Labs collected 363 survey responses from data practitioners and leaders across industries and regions between December 5, 2025 and February 1, 2026. Of the respondents, 73% identified as practitioners, while 27% serve in management or executive roles overseeing data teams.

Where applicable, results are compared to the 2025 State of Analytics Engineering survey to identify year-over-year shifts.

Percentages are reported as the share of respondents selecting each option, unless otherwise noted. For multi-select questions, respondents could choose more than one answer, so totals may exceed 100%.

In some charts, non-responses may be excluded from visualization for clarity, while percentages are still calculated as a share of total respondents to maintain consistency across the report.

Percentages are rounded to the nearest whole number for readability. As a result, sums of rounded values may differ from the rounded total by one percentage point.

Charts are generated from calculated survey values and then rounded for display. In some cases rounding individual components may produce totals that appear slightly different from rounded aggregate figures.

The 2026 respondent base remains broadly consistent with 2025 across industries, compensation ranges, and role composition. Respondents skew toward experienced, mid-to-senior data professionals, reinforcing that the findings reflect operational and strategic decision-makers.

Survey respondent profile

Analytics engineering continues to expand across industries, regions, and organizational contexts.

Respondents to the 2026 survey span a wide range of industries, including technology, financial services, healthcare, retail, and other sectors where compliance, governance, and data reliability are critical. Technology remains strongly represented, while participation across regulated and operationally complex industries reinforces the increasingly central role analytics engineering plays in enterprise environments.

Among survey respondents, participation remains concentrated in North America and Europe, reflecting both mature analytics ecosystems and the sourcing patterns of this year’s study.

Analytics engineers, data engineers, analysts, and data scientists continue to form the core professional composition of the field.

The field remains practitioner-led; 73% of respondents identify as practitioners, while 27% serve in management or executive roles.

Compensation signals continued demand for analytics expertise. Among practitioners in North America, more than 80% report earning over $100K. Manager-level compensation remains concentrated in higher salary bands, particularly in North America, while regional variation persists across Europe.

Despite increased AI integration and heightened strategic expectations, the daily work of analytics engineering remains grounded in maintenance and organization. A majority of respondents report spending most of their time maintaining or organizing datasets.

So while acceleration has reshaped workflows, it hasn’t yet displaced foundational data work.

What this means

Analytics engineering in 2026 is practitioner-led, enterprise-embedded, and operating at scale. The community reflects growing strategic responsibility, balancing innovation with operational accountability as AI integration accelerates and expectations rise.

And one more thing…

Even in an era defined by LLMs and governance frameworks, some debates remain timeless.

The data suggests that alignment on data governance may be easier than agreement on lunch.

Another question sparked strong opinions:

Responses reveal clear camps, and little consensus. Governance frameworks, it turns out, vary at 30,000 feet.

Behind every dashboard and pipeline is a community navigating complexity together. And if 2026 proves anything, it’s this: The future of analytics engineering will not be defined by tools alone, but by the people who know when to trust them, and when to verify.

Analytics engineering is moving fast and the teams pulling ahead are acting on these insights now. Save your seat for the 2026 State of Analytics Engineering virtual event on April 29 or 30 to hear how Katie Bauer, Jay Sobel, and Jason Ganz are putting this into practice. Get a clear view of what’s working across AI, governance, and cost control and come ready with your questions (plus a chance to win a few fun giveaways).