How We Use AI To Run A 90-Day Growth Audit

Most progress audits are a efficiency. Somebody exhibits up with a slide deck, interviews a number of stakeholders, and delivers a 40-page PDF that lives in a drawer. The group feels busy for 3 weeks, and nothing adjustments. I’ve been on each side of that transaction, and I received uninterested in it.

At my progress consultancy, we run 90-day progress sprints for venture-backed and personal fairness (PE)-backed corporations. The audit is the primary part. It used to take two to 3 weeks of guide work simply to get a transparent image of what was taking place inside an organization’s advertising and marketing group. Now, with AI woven into every step, we compress that discovery into days and spend the remaining time truly fixing issues.

Right here’s precisely how we do it.

Why Conventional Progress Audits Fail

The traditional consulting audit has a structural downside. The folks conducting it are incentivized to seek out complexity as a result of complexity justifies a much bigger engagement. So the deliverable turns into a laundry record of every thing that may very well be improved, ranked by nothing specifically, with no connection to what the enterprise truly wants within the subsequent quarter.

I ran advertising and marketing at corporations starting from Fortune 200 to early-stage startups earlier than beginning my very own agency. At one firm, a 30-minute assembly with the CEO required two or three pre-meetings simply to shine the deck. The choice was made in minutes. The deck went right into a drawer. All these hours, gone.

That have formed how I take into consideration audits. The output must be a working doc that turns into the blueprint for what occurs subsequent. Not a memento.

The AI-Assisted Audit Framework

Our audit covers three areas: the advertising and marketing org itself, the tech stack, and what I name AI readiness. That final one didn’t exist two years in the past. Now it’s arguably a very powerful piece, as a result of it determines how a lot of the roadmap an organization can truly execute with out hiring 5 extra folks.

Every space follows a particular course of, and AI exhibits up otherwise in each.

Part 1: Consumption And Context Constructing

Earlier than we discuss to anybody on the shopper’s group, we feed every thing we are able to get our palms on into Claude. Investor decks. Board shows. The corporate’s public advertising and marketing. Competitor inventive. Job postings from the final six months. Glassdoor evaluations. Product screenshots. Pricing pages.

Two years in the past, synthesizing all of that required a senior strategist spending a full week studying, annotating, and constructing a briefing doc. Now, we construct a complete context bundle in a day. Claude processes the uncooked materials and produces a structured temporary that features the corporate’s positioning gaps, messaging inconsistencies throughout channels, aggressive white area, and the questions we must be asking in stakeholder interviews.

The output isn’t a abstract. It’s a diagnostic framework tailor-made to that particular firm. We evaluation it, problem it, add our personal operator instincts, and stroll into discovery calls with a perspective as a substitute of a clean notepad. That adjustments the dialog instantly. Shoppers discover whenever you’ve performed the homework.

Part 2: Tech Stack And Workflow Mapping

That is the place issues get particular. We pull a full stock of all the instruments the advertising and marketing group makes use of. Buyer relationship administration (CRM). Electronic mail platform. Analytics. Attribution. Advert platforms. Content material administration. Design instruments. Undertaking administration. The typical mid-stage startup has between 15 and 30 marketing tools, and in virtually each audit, at the least a 3rd of them overlap or go largely unused.

We doc each workflow: how a marketing campaign goes from thought to reside, how leads get routed, how reporting occurs, who touches what, and when. Then we map every workflow towards what’s now potential with AI-native alternate options.

An actual instance: One shopper had three folks spending a mixed 40 hours per week on inventive manufacturing for paid social. Briefing a designer. Ready for rounds of revisions. Resizing for various placements. Exporting. Importing. We changed that workflow with a mix of AI inventive instruments and a customized automation that dealt with asset technology, versioning, and platform-specific formatting. The identical quantity of inventive now takes roughly eight hours of human time per week, and most of that’s strategic evaluation slightly than manufacturing.

Instruments like HeyGen and ElevenLabs deal with video and audio manufacturing that used to require a studio. Customized AI brokers constructed on open-source AI harnesses like OpenClaw and Hermes automate analysis, aggressive monitoring, and content material drafts. The purpose isn’t to name-drop software program. It’s that the panorama of what may be automated has expanded dramatically within the final 18 months, and most advertising and marketing groups haven’t caught up.

Part 3: AI Readiness Evaluation

This part is the one which surprises shoppers essentially the most, as a result of it’s much less about expertise and extra about folks.

We consider three issues. First, does the group have the curiosity and willingness to undertake AI instruments? Some groups are keen. Some are terrified. Figuring out the place folks stand earlier than you begin pushing new workflows prevents the kind of resistance that kills transformation projects. I spoke about AI readiness to a gaggle of senior entrepreneurs at a hyper-growth client app, and the primary query requested was: “Isn’t the magic in our human work and interactions?” They had been afraid.

Second, does the corporate’s information infrastructure truly help AI-driven optimization? In case your CRM is a multitude, your attribution is broken, and your analytics are constructed on self-importance metrics, no AI instrument goes to avoid wasting you. Rubbish in, rubbish out nonetheless applies. We flag the info hygiene points that have to be fastened earlier than any AI implementation will produce dependable outcomes. And the audit acknowledges the info gaps and the way (and why) to repair them.

Third, the place are the highest-leverage automation alternatives? Not every thing must be automated. Inventive technique nonetheless requires human judgment. Model selections nonetheless want a human with style and context. The audit identifies which workflows will profit most from AI and which of them want a human firmly within the loop. AI readiness isn’t about changing all people with AI instruments and brokers.

What The Deliverable Truly Appears to be like Like

We don’t hand over a deck. We produce a shared doc with 4 sections: present state analysis, prioritized alternative map, 90-day implementation roadmap, and a tool-by-tool advice record with estimated time and price financial savings.

The roadmap breaks the 90 days into three phases. The primary month focuses on fast wins, the workflows the place AI may be plugged in with minimal disruption and fast impression. Month two tackles the structural adjustments, issues like rebuilding attribution fashions or redesigning the content material manufacturing pipeline. Month three is about coaching and handoffs, making certain the group can run the brand new programs independently.

The doc is collaborative. Shoppers can remark, push again, and reprioritize. It turns into the working blueprint for the engagement, not a PDF that will get emailed and forgotten.

The place The Actual Financial savings Present Up

The financial savings are hardly ever the place folks anticipate them. Most founders assume AI will minimize their advert spend or cut back their company charges. Typically it does. However the greater wins are usually in time recaptured.

A advertising and marketing group that was spending 60% of its week on manufacturing and reporting and 40% on technique will get these numbers flipped. People deal with the work that really requires style, judgment, and relationship-building. The AI handles the repetitive execution that was consuming their calendars.

One engagement decreased a shopper’s inventive manufacturing cycle from three weeks to 4 days. One other automated their weekly reporting entirely, liberating up a senior analyst to deal with precise evaluation as a substitute of pulling numbers into slides. A 3rd rebuilt their electronic mail lifecycle from scratch utilizing AI-generated segmentation and content material, which minimize their price per acquisition by 30% within the first 60 days.

None of these outcomes required firing anybody. They required shifting folks from low-leverage duties to high-leverage duties. That’s the a part of the AI dialog that gets lost in the layoff headlines.

What I’d Inform Any Advertising Chief Studying This

You don’t want to rent a agency to start out. Decide one workflow on your team that is repetitive, time-consuming, and doesn’t require deep inventive judgment. Map it out step-by-step. Then ask whether or not an AI instrument might deal with any of these steps right this moment.

Start by tackling reporting. Subsequent, deal with aggressive analysis. Contemplate first-draft content material manufacturing as an early win. Lastly, provoke the method wherever the ache is loudest and the chance is lowest. Get a win. Present the group what’s potential. Then broaden.

The businesses that can wrestle are those ready for somebody handy them a playbook. The businesses that can win are those operating their very own experiments proper now, even clumsy ones, and studying what works inside their particular context.

The audit is only a structured method to do what each advertising and marketing group ought to already be doing: trying actually at how time will get spent and asking whether or not there’s a greater means. AI simply made “higher” much more accessible than it was 18 months in the past.

Extra Sources:

Featured Picture: Tetiana Yurchenko/Shutterstock