Why faster AI drafting may not speed marketing delivery

AI can produce a first draft in seconds. Knak’s Marketing Production in the Age of AI report says 85% of surveyed marketing teams missed at least one campaign launch date over the last year because of workflow constraints. That gap matters to CEOs and marketing leaders because drafting speed is only one component of productivity. If faster generation creates more editing, choices, handoffs, and approvals, the saved time may never become usable capacity.

Faster content generation can leave total production time unchanged

Generative AI can cut the time needed to create draft marketing material. A team can quickly turn an idea into email copy, subject lines, images, and alternative treatments. The management question comes next: how much sooner does the finished campaign launch?

A draft still has to fit the campaign strategy, customer need, brand voice, and offer. It then moves through design, proofing, testing, approval, scheduling, and coordination. Faster drafting improves one stage of a larger system. The business result depends on total effort and end-to-end cycle time.

This is a bounded argument about workflow. Knak also reports that some teams using AI complete work substantially faster, so AI use and higher productivity can coexist. Executives need to distinguish between automating a task and increasing the throughput of the system around it.

AI is accelerating one part of the production workflow

Knak reports substantial AI use across several marketing-production activities. Knak sells marketing-production technology, so it has a commercial interest in how companies understand and address production workflow. Its survey findings are therefore vendor-sponsored evidence about reported practices.

Knak places the largest reported delays in other parts of the workflow. Faster copy generation will have limited effect on launch time when campaigns spend substantial time waiting for creative work, coordination, or decisions.

Knak also reports that 60% of teams need at least four people to produce a single email. More participants can create more handoffs, dependencies, and decisions. Drafting automation reduces those costs only when it changes the surrounding workflow as well.

The same issue appears in Knak’s editing figures. It reports that 70% of teams have deployed AI within marketing production, while 88% say AI output still requires moderate or substantial human editing. These figures show AI adoption alongside human editing, but they do not establish that AI causes additional editing.

Human review has substantive purposes. Marketing teams must determine whether a message fits the objective, brand, offer, and customer, while also assessing accuracy and persuasiveness. Faster generation lowers the effort needed to create a first draft. Its effect on total review effort depends on how the team uses the output.

Cheap generation changes the economics of variation

Generative AI makes additional versions inexpensive to create. A team that once prepared two subject lines can quickly prepare 10, or produce several openings and calls to action before choosing a direction. That capacity can help when variants support a defined test or explore materially different approaches. It can also give people more material to evaluate.

Knak reports that 69% of teams go through two or three rounds of revisions before receiving approval. In that workflow, the cost of a new variant includes human review. Everyone asked to compare, edit, or approve an option adds time beyond the near-zero marginal cost of machine generation.

The management issue is the purpose of variation. Clear hypotheses can make alternatives valuable, while unbounded generation can expand the set of decisions a team must make before publication. Leaders should test whether additional output raises campaign throughput or mainly increases selection and revision work.

AI can also produce concrete material before a team has settled the audience, objective, offer, or message. Revision may then carry strategic decisions that could have been made before drafting. Later changes to the offer or audience emphasis can alter work already produced. Leaders should test this workflow risk inside their organizations; Knak’s survey figures do not establish it as a causal result.

Saved production time can be absorbed by higher production demand

The larger promise of AI is organizational capacity. Knak reports that 82% of respondents spend at least half their time on production rather than planning or strategy. Combined with the reported campaign delays, this points to substantial production pressure among the survey respondents.

One possible explanation is that faster generation raises expected output. An organization may respond to lower drafting costs by requesting more campaigns, channels, or variants, absorbing the hours saved at one stage with additional production. Leaders can test this hypothesis against their own operating data.

Other explanations remain possible. Approval structures, revision practices, production requirements, or broader organizational conditions may shape both workload and AI use. The reported associations do not establish which factor causes another. Executives need end-to-end measures before treating faster task completion as evidence of added strategic capacity.

A practical test is to follow the saved hours. When one stage becomes faster, leaders can measure whether total production effort falls, launch cycles shorten, or planning time rises. That shows whether a local efficiency gain creates capacity for the organization or is absorbed elsewhere in the workflow.

The faster teams point to workflow design

Knak reports that teams able to complete an email in around four hours tend to involve fewer people and use AI deliberately rather than experimentally. This provides a useful counterpoint: Knak’s reported data shows an association between AI use and fast production.

The association shifts attention toward workflow design. Fewer participants can reduce handoffs and approval dependencies, while deliberate AI use can focus automation on defined tasks. Those practices offer a plausible explanation for faster production, although the reported association alone cannot establish causality.

Already-efficient teams may have processes that influence both production speed and how they adopt AI. Leaders can treat the finding as a prompt for investigation. Comparing participant counts, revision rounds, approval time, and AI use across internal teams can show which conditions accompany faster delivery.

Measure throughput before declaring an AI productivity win

AI productivity should be measured where the organization expects a business benefit. If the objective is faster campaign delivery and greater strategic capacity, drafting time is only one measure. Leaders also need visibility into revision rounds, human editing, approval time, handoffs, participant count, total production effort, and the full cycle from an agreed campaign direction to launch.

This changes how an AI deployment is evaluated. A team may reduce first-draft time while adding variants, review work, or approval activity, leaving overall throughput unchanged. Another may use AI while reducing total production effort or cycle time. That result provides stronger evidence of usable capacity.

Measurement can also guide workflow changes. When approval waiting time dominates the cycle, leaders can examine approval authority and participant count. When revision dominates, they can test whether objectives, requirements, and brand guidance are settled before generation. When repeated construction consumes time, they can measure whether templates or modular production reduce that burden.

Knak Chief Marketing Officer Jennifer Delevante observes in the study that marketers were promised AI would give them back time for strategy, while the data suggests that time has not materialized yet. As an executive at Knak, which has a commercial stake in improving marketing-production workflows, Delevante is interpreting data in a market relevant to Knak’s business. The operational question for leaders is whether cheaper generation leaves the whole workflow with more usable capacity.

Key takeaways for leaders

• Measure end-to-end production time: Faster AI drafting creates value when it reduces total campaign effort or launch time. Marketing leaders can track the full cycle from agreed direction through approval and launch.
• Target the workflow bottlenecks: Approvals, creative production, coordination, and human editing can absorb gains from faster generation. Marketing operations teams can identify where campaigns wait longest and redesign those stages first.
• Control the cost of variation: Cheap AI generation makes it easy to produce more options, while every additional option may require human review. Campaign owners can tie variants to defined hypotheses and measure whether they improve outcomes enough to justify the review work.
• Follow the saved hours: Lower drafting costs may lead organizations to request more campaigns, channels, and variants. Marketing leaders can track whether AI savings reduce production effort, shorten launch cycles, or increase time available for planning and strategy.
• Design AI around efficient workflows: Knak associates faster email production with fewer participants and deliberate AI use, although its data does not establish causation. Marketing operations teams can compare participant counts, revision rounds, approval times, and AI practices across internal teams.
• Measure AI productivity through business outcomes: Drafting speed alone cannot show whether AI creates usable capacity. Executives evaluating AI investments can track total production effort, revision rounds, handoffs, approval time, and end-to-end cycle time.