sharkitect-solutions/executing-marketing-campaigns

Executing Marketing Campaigns

Campaign Type Decision Tree

Before doing anything, classify the request. The campaign type determines every downstream decision -- framework, channel mix, timeline, and metrics.

What outcome does the user need?

  Generate pipeline / leads        --> DEMAND GEN       (references/campaign-frameworks.md #demand-gen)
  Launch a product or feature      --> LAUNCH           (references/campaign-frameworks.md #launch)
  Reduce churn / increase LTV      --> RETENTION        (references/campaign-frameworks.md #retention)
  Build brand awareness / category --> AWARENESS         (references/campaign-frameworks.md #awareness)
  Re-engage lapsed users           --> WIN-BACK         (references/campaign-frameworks.md #win-back)
  Test a new channel or message    --> EXPERIMENT       (references/campaign-frameworks.md #experiment)
  Multiple / unclear               --> Ask. Do NOT guess. Wrong type = wasted budget.

After classifying, pull the matching framework from the reference file. Each framework includes timeline, channel mix, budget split, success metrics, and kill criteria. Do not improvise these -- use the framework.

Expert Rules (What Generic Advice Gets Wrong)

These are the practitioner insights that separate campaigns that work from campaigns that "looked good on paper."

1. Measurement-First Principle

Set up tracking, attribution, and reporting BEFORE writing a single line of copy. If you cannot measure the outcome before launch, do not launch. This is the single highest-leverage rule in campaign execution.

• See references/measurement-frameworks.md for attribution setup.

2. Channel Selection > Creative Quality

The right message in the wrong channel produces zero results. The mediocre message in the right channel still generates pipeline. Always validate channel-audience fit before investing in creative polish.

• Channel selection guide: references/channel-playbooks.md

3. The 70/20/10 Budget Rule

• 70% to proven channels (known positive ROI from past data)
• 20% to testing (channels with signal but unproven at scale)
• 10% to experimental (new channels, no data yet) Deviation requires explicit justification with data.

4. Pre-Launch Validation Kills "Launch and Hope"

Most campaigns fail because of no pre-launch validation, not bad execution. Before full launch, always run:

• Message testing (3-5 variants, small audience, 48-72 hours)
• Landing page smoke test (paid traffic to measure conversion before scale)
• Internal dry run (full sequence walkthrough to catch broken links, wrong UTMs, missing tracking pixels)

5. Diminishing Returns Curve

Every channel has a saturation point. Signs you have hit it:

• CPA rising 20%+ week-over-week with no audience expansion
• Frequency > 3x per user per week on paid social
• Email list engagement dropping below 15% open rate When you see these: stop scaling that channel. Diversify.

6. Cadence > Perfection

One consistent campaign per month outperforms one "perfect" campaign per quarter. The compounding effect of regular audience touchpoints beats sporadic brilliance every time. Build a campaign calendar, not a magnum opus.

Anti-Pattern Decision Table

| Anti-Pattern | Why It Fails | What To Do Instead | |---|---|---| | Launching without measurement in place | Cannot prove ROI, cannot optimize, budget gets cut next quarter | Set up tracking + attribution first (references/measurement-frameworks.md) | | Spreading budget equally across all channels | Ignores channel-audience fit; subsidizes underperformers | Use 70/20/10 rule; allocate by proven performance data | | Optimizing for clicks/impressions instead of pipeline | Vanity metrics look good in reports but do not generate revenue | Define business metric (MQLs, SQLs, revenue) as primary KPI from day one | | Running campaigns without audience segmentation | Same message to everyone = resonates with no one | Segment by pain point, buying stage, or persona before writing copy | | "Always be launching" -- too-frequent sends | Campaign fatigue tanks engagement rates across all channels | Cap email to 2-3/week max; paid social frequency < 3x/user/week | | No kill criteria before launch | Failing campaigns run too long, burning budget on hope | Define kill threshold at launch (e.g., "if CPA > $X after 500 clicks, pause") | | Copying competitor campaigns without context | Their audience, budget, and goals are different from yours | Analyze competitor strategy; adapt the principle, not the execution | | Treating channels as independent silos | Ignores cross-channel amplification and attribution overlap | Map the customer journey across channels; use multi-touch attribution | | Skipping the post-mortem | Same mistakes repeat; no institutional learning | Run a structured post-mortem within 1 week of campaign end (references/measurement-frameworks.md #post-mortem) |

Rationalization Table

| The User Says | They Actually Need | Why | |---|---|---| | "We need a viral campaign" | Awareness campaign with realistic reach targets | Virality is an outcome, not a strategy. Build for consistent reach. | | "Let's do everything -- email, social, paid, events" | Channel prioritization based on audience data | Spreading thin = mediocre everywhere. Pick 2-3 channels max to start. | | "Our campaign isn't working" | Campaign health diagnostic (see below) | "Not working" has 4 distinct failure modes. Diagnose before prescribing. | | "We need more leads" | Demand gen campaign with pipeline metrics | "More leads" often means "more qualified leads." Define MQL criteria first. | | "Competitor X is doing Y, we should too" | Competitive analysis then adapted strategy | Copying tactics without understanding strategy wastes budget. | | "We just need better creative" | Full-funnel diagnosis | Creative is rarely the bottleneck. Usually it is targeting or channel fit. | | "Can we just boost this post?" | Paid social strategy with targeting and measurement | Boosting without targeting burns budget. Set up proper paid campaigns. |

NEVER List

• NEVER launch a campaign without measurement and attribution configured first
• NEVER allocate budget without historical channel performance data or a test plan
• NEVER optimize toward vanity metrics (impressions, likes) as the primary KPI
• NEVER skip audience segmentation -- "everyone" is not a target audience
• NEVER run a campaign longer than 2 weeks past its kill criteria threshold
• NEVER copy a competitor's campaign without analyzing why it works for their audience
• NEVER send more than 3 marketing emails per week to the same segment
• NEVER present campaign results without comparing to pre-defined success metrics

Campaign Health Diagnostic

When a user says a campaign is "not working," diagnose which stage is broken:

Funnel Stage     | Symptom                  | Root Cause Area        | Fix Direction
-----------------+--------------------------+------------------------+---------------------------
Awareness        | Low impressions/reach    | Channel or targeting   | Expand audience, test new
                 |                          |                        | channels, increase budget
Interest         | Low CTR (< 1% display,   | Creative or messaging  | Test new headlines, value
                 | < 2% email, < 3% search)|                        | props, or CTAs
Conversion       | Low conversion rate      | Offer or landing page  | Simplify page, strengthen
                 | (< 2% landing page)     |                        | offer, reduce form fields
Retention        | High churn post-convert  | Expectation mismatch   | Align campaign promise with
                 |                          |                        | actual product experience

Work top-down. An awareness problem makes everything downstream look broken. Fix the highest stage first.

Workflow: Campaign Build Sequence

After classifying campaign type and pulling the framework:

1. Measurement setup -- tracking, UTMs, attribution model, reporting dashboard
2. Audience definition -- segments, personas, exclusions
3. Channel selection -- pick 2-3 based on audience data + 70/20/10 budget rule
4. Message development -- per-channel messaging, value props, CTAs
5. Pre-launch validation -- message test, landing page smoke test, internal dry run
6. Launch -- staggered rollout, monitor first 48 hours closely
7. Optimization -- weekly review cadence, reallocate budget based on data
8. Post-mortem -- within 1 week of campaign end, structured review

Each step has detailed checklists in the reference files.

References

• references/campaign-frameworks.md -- Campaign architecture by type (launch, demand gen, retention, awareness, win-back, experiment)
• references/channel-playbooks.md -- Per-channel execution checklists and optimization guides
• references/measurement-frameworks.md -- Attribution models, metric hierarchies, reporting templates, post-mortem framework