The Tool Stack

Specific vendor names below are current as of this research (August 2026); pricing, feature sets, and even which vendors still exist move fast in this category — verify current offerings directly before budgeting around any specific tool.

---

#The execution layer

Google Ads, Meta Ads Manager, and TikTok Ads are the three platforms almost every managed-media account touches in some combination — this course assumes familiarity with running campaigns on at least the first two as a baseline competency, not something taught here. For clients with larger, more diversified budgets, programmatic buying through a demand-side platform (DSP) — The Trade Desk and Google's DV360 are the two most commonly cited — extends reach into display, video, and connected-TV inventory that the walled-garden platforms above don't cover. Running a DSP well is a materially different skill from running Meta or Google campaigns and is generally a later-hire specialization (Module 5), not a day-one requirement.

#The reporting and attribution layer

This is the layer that turns raw platform data into the client-facing report that actually justifies the retainer, and the right tool depends on the client's business model, not a single "best" answer:

• Triple Whale and Northbeam are built specifically for DTC/e-commerce accounts, covering pixel-based attribution, marketing-mix modeling (MMM), and incrementality testing in one platform. [Directional] For a book of business concentrated in DTC e-commerce, one of these two is the standard starting point.
• Supermetrics and Funnel.io solve a different problem: pulling raw platform data into a client's own existing BI stack (Looker, Tableau, Power BI) rather than replacing it with a new dashboard. [Directional] These fit better when the bottleneck is data-pipeline plumbing, not attribution modeling itself, or when the client already has (and wants to keep) their own analytics stack.
• A statistical-incrementality tool becomes worth the added cost once a client's monthly media mix crosses roughly $50,000/month — below that, the added precision rarely changes a real budget decision enough to justify the tool's cost and setup complexity. [Speculative] — this specific breakpoint is the kind of number incrementality-tooling vendors themselves are positioned to want low (a lower stated threshold means more agencies conclude they need the product sooner), and this course could not trace it to an independent, vendor-neutral source; treat it as a rough rule of thumb worth testing against your own client's actual budget-decision-making, not an audited threshold.

#Billing infrastructure

Module 3 covered the "playing the bank" mechanism — fronting client ad spend on the agency's own card and invoicing later. The operational tooling that supports this safely, rather than through a founder's personal card: virtual-card platforms built for agencies (Pliant is one named example) that issue per-client or per-campaign virtual cards with spend limits, giving the agency billing control and float without exposing one shared card number across every client account, and making it easier to trace exactly which client's spend caused which charge. [Directional]

#Project management and client communication

Nothing in this category is specific to advertising agencies — Asana, ClickUp, and Monday.com for internal workflow, Slack or a client-facing Slack Connect channel for day-to-day communication, and a proposal tool (rather than a static PDF, per Module 6's finding that PDF attachments frequently go unopened — flagged there as [Speculative] on the specific percentage, since it's a statistic proposal-software vendors have a direct commercial interest in) round out the standard stack. This is genuinely commodity infrastructure; the specific vendor matters far less here than in the execution and reporting layers above.

#Where AI tooling has actually changed the job, not just the marketing around it

This deserves its own note because it's easy to either dismiss as hype or overstate as replacement, and the honest 2026 picture is neither:

Meta's Advantage+ and Google's Performance Max now handle the tactical layer — bid adjustment, audience targeting, creative rotation, budget pacing — that used to be manually managed. Teams report a 60–87% reduction in time spent on tactical optimization once these tools handle that layer. [Directional], with a caveat worth stating plainly: this course could not trace the figure to a specific named study, and a wide range like 60–87% is itself a sign of loosely-comparable underlying measurements rather than one tight finding — treat the direction (real, substantial time savings) as more solid than the precise range, in the same way SMMA's platform-shifts lesson treats Meta's own self-reported Advantage+ performance numbers with caution, since a platform (or a tool vendor built around AI-driven campaign automation) reporting how much time its own product saves carries the same structural incentive to report favorably. This is covered in depth in Module 9 as a structural threat to the traditional media-buyer value proposition; the tools section here is just the practical note that if you are not using these AI-driven campaign types by default in 2026, you are running a materially more expensive and more labor-intensive operation than a competitor who is — this is now closer to table stakes than a differentiator.

Adoption is real but not universal: only 29.1% of agencies report using AI for media planning and 22.1% for media-buying strategy specifically, despite both being high-leverage use cases — meaning a meaningful share of the industry is still running the old, fully-manual process. [Directional] This gap is itself informative: it suggests genuine competitive room for a new agency built AI-native from day one against slower-moving incumbents, which is the framing Module 9 develops further.