Rules
Part of A 90-day plan to set up marketing operations in the right order
5 parts of marketing operations case studies, explained
Five parts of marketing operations case studies: the vendor story, the baseline, the mechanism, the missing evidence, and the local proof test.
What to take away
- Five parts of marketing operations case studiesthe vendor story, the baseline, the mechanism, the missing evidence, and the local proof test.
- Five vendor-published accounts to read them againstGannett with Asana, Genpact and Cisco with Adobe Workfront, Canva with monday.com, and Siemens with Wrike.
- Gannett and Asana report 120 to 150 monthly campaigns and 389 workdays saved; Genpact and Adobe Workfront report connected intake through publishing.
- Every one of the five is a vendor's account of a customer, so the figures describe one selected context rather than a repeatable result.
- Ask for the formula, the period, adoption, costs and failed iterations before you compare a figure to your own queue.
- Transfer one mechanism, run it on a limited queue, and keep the baseline so you can judge it.
Marketing operations teams read case studies to find mechanisms, not to copy outcomes. A vendor story is a selected account of one customer, published by the company that sold the software. That does not make it false. It makes it incomplete in ways the page rarely states.
Part 1: the vendor story
The first part is who published the account and who it describes. Adobe published the Genpact story in June 2026, and Asana published the Gannett account. Five named accounts show the pattern, and each one names a publisher with a product to sell.
| Published account | Publisher | Why a buyer would read it |
|---|---|---|
| Gannett | Asana | High campaign volume inside a large media group |
| Genpact | Adobe Workfront | Services firm, intake through publishing |
| Cisco | Adobe Workfront | Enterprise intake for marketing and creative work |
| Canva | monday.com | In-house design and marketing teams |
| Siemens | Wrike | Marketing work spread across regions |
List prices for these tools are public as of 2026. Asana lists Starter at about $10.99 per user per month and Advanced at about $24.99, billed annually. monday.com Work Management runs from about $9 to about $19 per seat per month billed annually.
Adobe sells Workfront in Select, Prime and Ultimate editions, with pricing quoted on request. Wrike lists Team at about $9.80 per user per month and Business at about $24.80.
Record the relationship before the numbers. Gannett publishes USA TODAY, Genpact sells professional services, Cisco sells networking equipment, Canva sells design software, and Siemens is an industrial group. A named customer with a logo, a quote and a linked product page is marketing material, and the selection is the publisher's.
Part 2: the baseline
The second part is the starting point the result is measured against. Asana reports Gannett moving from 50 to 70 monthly campaigns to 120 to 150, with 389 workdays saved.
That range is a definition problem before it is a precision problem. Ask what counted as a campaign, over which period, and who counted it. The same three questions apply to the Genpact, Cisco, Canva and Siemens accounts.
Part 3: the mechanism
The third part is the thing that actually changed. Asana describes centralized requests and execution through forms, templates, rules, and dependencies. The same account covers portfolios, reporting, and workload views.
Adobe describes Genpact running Workfront as an operational backbone. The work includes campaign intake, creative briefs, and channel briefs. Reviews, approvals, schedules, and Salesforce campaign records sit in the same system.
The transferable idea in the two detailed accounts is a visible operating record replacing scattered requests. That is the same intake and capacity question that runs through common marketing operations questions.
Part 4: the missing evidence
The fourth part is what the page does not say. Neither the Gannett nor the Genpact story publishes the formula behind its figures, the adoption curve, the implementation cost, or the iterations that failed.
Records to request before the demo
- Baselinedefinition, count, period
- Mechanismworkflow, adoption, controls
- Outcomedenominator and full results
- Costimplementation and operation
- References at similar scale and risk
Build the request before the demo. Ask for the metric definition and period, the work population, and the team and training load. Also ask for the integrations and who owns them, plus the concurrent changes and the customer-facing effects.
| Claim element | Record to request | Risk if missing |
|---|---|---|
| Selection | Why this customer, and why now | A best case reads as typical |
| Baseline | Definition, count, period | Change is not comparable |
| Mechanism | Workflow, adoption, controls | Tool gets all credit |
| Outcome | Denominator and full results | Selected success dominates |
| Cost | Implementation and operation | Benefit is overstated |
A tool supports a process; it does not create clarity by itself. Ask for references at similar scale and risk, and read the current capabilities rather than the story's version of them.
Part 5: the local proof test
The fifth part is your own test. Pick one mechanism, such as structured intake or capacity visibility, and run it on a limited queue.
Keep the baseline, the operator effort, the defects, the exceptions and the customer guardrails. Compare like with like, and review the failures before you scale.
Case studies support discovery. Proof testing, security review and contracts should govern the purchase. That is also why marketing operations benchmarks need a stated basis before anyone quotes them.
Audit the published evidence
The FTC advertising substantiation policy requires a reasonable basis for objective advertising claims before they are disseminated. Whether a specific performance claim meets that standard depends on the facts, so take advice on your own.
The FTC guidance for marketers using reviews covers fake feedback, selective requests, conditioned incentives and undisclosed relationships. Apply the same checks when a case study page names providers.
Common questions
Are vendor case studies reliable?
They can report a customer's account accurately. Selection, the commercial relationship and omitted evidence limit what another buyer should infer from them.
What should a buyer verify first?
Start with the formula, the baseline, the period and the work population. Then ask about adoption, concurrent changes, implementation effort and total cost.
How should a case-study idea be tested?
Pick one mechanism, define the expected change and the guardrails, and keep a baseline. Run representative work, then review failures and exceptions before scaling.
