2 minutes

Part 5: Know What's Actually Possible

The Problem

Your developer says: “That integration will take 8 weeks.”

Is that real? Or are they sandbagging because they don't want to do it? You can't tell. So you either push back blindly or accept it blindly. Both cost you money.

Timeline Reality Checks

Here's what common marketing requests actually take—and why.

RequestTimelineWhyRed Flag
Simple API Integration (pre-built connector, like Zapier)2-5 daysTesting + edge case handlingClaims "instant setup" but requires custom fields
Custom API Integration (no existing connector)3-6 weeksAPI documentation review, development, testing, security reviewVendor says "2 weeks" without asking about your security process
AI Chatbot (from template)1-2 weeksTraining data upload, response testing, edge case refinementClaims "live in 48 hours" but doesn't mention training phase
AI Chatbot (custom)6-12 weeksCustom training, integration with your knowledge base, approval workflowsVendor doesn't ask about your content sources
Data Migration (CRM, MAP, CDP)4-12 weeksData mapping, cleaning, validation, deduplication, testingVendor doesn't ask about data quality or volume
Landing Page Personalization (rules-based)1-2 weeksSegment definition, rule setup, QA testingClaims "real-time AI" but it's if/then rules
Landing Page Personalization (AI-driven)6-8 weeksModel training, A/B testing, learning periodVendor doesn't mention "learning phase"
Email Automation Workflow (simple)3-5 daysTrigger setup, template creation, testingNo mention of deliverability testing
Email Automation Workflow (complex, multi-touch)2-4 weeksLogic mapping, conditional branching, edge casesVendor doesn't ask about your current logic flows

What Slows Things Down (That No One Tells You)

Your IT Security Review

2-4 weeks

Any new tool that touches customer data must pass security review. If the vendor hasn't mentioned this, they don't know your process.

Questions to ask internally:

  • Does this tool require a security review?
  • How long does our security team take to approve new tools?

Data Quality Issues

1-8 weeks

AI/ML tools fail with dirty data. If you have duplicate records, missing fields, or inconsistent formatting, you'll spend weeks cleaning before the tool works.

Questions to ask internally:

  • What's the quality of our CRM data?
  • Do we have a data cleanliness score?

Approval Workflows

1-4 weeks

Most B2B companies require 3-5 stakeholders to approve new tools (IT, Security, Legal, Finance, Exec). Vendor timelines assume instant approval.

Questions to ask internally:

  • Who needs to approve this purchase?
  • What's the realistic timeline for approvals?

Change Management & Training

2-6 weeks

Your team needs to learn the new tool. If the vendor doesn't include training in the timeline, you'll spend weeks post-launch ramping up.

Questions to ask internally:

  • What does training look like?
  • How long until our team is self-sufficient?

“Can We Build This?” Quick Guide

Common requests and whether they're realistic:

Can AI write our weekly newsletter?

Yes, AI can draft it (Level 4 capability). But you'll need to edit for brand voice, fact-check, and approve.

Realistic timeline: 30-40% time savings, not 100% automation

Can we auto-personalize our website for every visitor in real-time?

Possible with right tech stack (CDP + personalization engine). Requires clean data, defined segments, and A/B testing period.

Cost: $50K-$200K/year depending on scale. Timeline: 3-6 months to see ROI

Can AI predict which leads will convert?

Yes, predictive lead scoring works (if you have enough data). Needs 1,000+ historical leads minimum for accuracy. Model retraining required quarterly.

Timeline: 2-3 months to train initial model + ongoing tuning

Can we integrate our CRM with our new marketing platform?

Yes, if both have APIs. May require middleware (like Zapier) if no direct connector.

Cost: $0 (Zapier) to $25K (custom integration). Timeline: 2 days (Zapier) to 6 weeks (custom)

Can AI create our social media content and auto-post it?

AI can draft posts (Level 4). Auto-posting without review = brand risk. AI doesn't understand trending topics or real-time context.

Use AI for drafts, human for approval/posting

Can we set up a chatbot that answers any customer question?

Yes, if you have a comprehensive knowledge base. Accuracy = 70-85% at best. Requires human fallback for complex questions.

Timeline: 4-8 weeks for training + ongoing refinement

What You Just Learned

  • You now know realistic timelines for common marketing requests
  • You understand what slows down implementations (and it's not always the vendor)
  • You can challenge developers and vendors with confidence

You've Completed The STACK Framework

Congratulations! You're now technically literate.

In 10 minutes, you learned:

  1. 1How to separate signal from noise with 3 questions
  2. 2How to translate 12 common technical terms
  3. 3What AI can (and can't) do across 5 capability levels
  4. 4The 7 red flags that signal vendor theater
  5. 5Realistic timelines and technical constraints

You're now equipped to:

  • Evaluate martech vendors without getting lost in jargon
  • Challenge unrealistic timelines from developers
  • Spot AI theater vs real capabilities
  • Make smarter budget decisions

What's Next?

Get Ongoing Intelligence

You just learned the foundations. Now stay sharp with daily briefings.

Join thousands of marketing leaders getting AI Ready CMO:

Free subscribers get:

  • Daily intelligence briefings (5-min read)
  • Access to workshops, resources, and courses

Pro subscribers get:

  • Monthly trend reports
  • Weekend deep-dives
  • 6 proprietary tools

Free to learn. Paid to lead.