Forrester: Your AI Productivity Gains Might Be Creating More Work For Everyone Else

Forrester Principal Analyst Lisa Gately warned this week that individual AI productivity gains are quietly making organizations less efficient overall. Citing Shopify CEO Tobi Lütke’s warning against tossing “AI slop grenades” — using AI to generate work quickly and shifting the burden of fixing it onto colleagues — Gately’s math is blunt: if AI saves one employee 20 minutes but creates 30 minutes of downstream cleanup for someone else, that is not a net productivity win.

Drawing on her new report, “The State Of Content Creation And Optimization Solutions, 2026,” Gately found 66% of marketing leaders say their organizations are still in an experimentation phase with AI for content and creative work — meaning most AI-assisted content workflows still lack the norms that would prevent the cleanup-cost problem she’s describing. Her recommendations: set explicit rules for when AI should be used and what quality bar its output has to clear before a human sees it, require mandatory review at named stages with named accountability, and measure AI’s impact across the entire workflow rather than the individual time it saved one person. She also flags that 51% of marketing leaders report reducing reliance on outside agencies as AI adoption grows — the labor isn’t disappearing, it’s shifting in-house and often ungoverned.

That caution lands directly against the sales-tech pitch. A mid-2026 piece from Highspot argued enterprise sales teams need well-chosen AI productivity tools rather than scattered point solutions, explicitly warning that layering poorly-fit AI tools onto an existing stack creates its own sprawl. Same underlying claim from opposite directions: adding AI tools without deliberate workflow design is the failure mode, whether you’re buying the tools or governing what they produce.

For revenue teams building AI-assisted prospecting, content, or enablement workflows, Gately’s piece is a direct check against declaring victory on tool rollout before measuring true end-to-end throughput. The AI-in-sales stack keeps expanding; the harder work no tool automates is deciding which parts of it earn a permanent place in the workflow, and which just move the cleanup cost downstream to whoever inherits the output next.