Agency Selection Scorecard

Picking an agency is not a gut call. Treat it like hiring a senior leader. Below is a practical scorecard you can run quickly and repeatably.

Core evaluation pillars (expertise, fit, tech)

• Expertise: Do they have repeatable playbooks for your problem? Look for case studies with similar deal sizes, sales cycles, and buying committees.
• Fit: Cultural and process fit matter more than shiny credentials. Do their rhythms match yours? Can they work during your core hours?
• Tech: Can they operate within your stack? Ask for proof of ownership of the tools they propose, plus admin-level access patterns.

Score each pillar 1-10. Use notes to justify scores. If any pillar is below 5, walk away or change scope.

Weighted scoring template (example weights)

Use weights to reflect what really matters for you. Example:

• Expertise 40%
• Fit 30%
• Tech/ops 20%
• Commercial terms 10%

Calculate: final score = sum(weight * score). Normalize to 100. Use thresholds: >75 green, 60-75 amber, <60 red.

Evidence requirements (metrics, references, org chart)

Ask for hard evidence, not marketing slides:

• Metrics: pipeline generated, average deal size supported, conversion rates by funnel stage, payback period of campaigns.
• References: three clients in region or vertical. Ask for CMO or head of revenue, not a marketing manager.
• Org chart: who will do the work? Get names, titles, local hires versus remote resources, and estimated FTE allocation.

If they push back on sharing org chart or raw metrics, that tells you more than their slides.

Interview questions to validate capabilities

Ask the people who will run your account, not just sales.

• "Show me the last campaign you ran that failed. What did you learn and what did you change?"
• "Walk me through a lead from first touch to accepted opportunity in your system. Where do you hand off to sales?"
• "What single metric would you target first with us, and why?"
• "If our first quarter underperforms, what three corrective actions will you take?"
• "Which tools in our stack would you replace and why?"

Listen for specifics. If answers are generic, the work will be too.

Engagement Models and Pricing

Stop thinking in fixed models. Pick the model that aligns incentives and risk.

Common engagement types (retainer, project, hybrid)

• Retainer: ongoing strategic work and operations. Good when you need continuity.
• Project: fixed-scope launches, redesigns, migrations. Use when outcomes are well-defined.
• Hybrid: retainer plus performance or milestone bonuses. Use this when you want accountability on specific KPIs.

Pricing benchmarks by service and scope

Benchmarks are blunt instruments but handy for sanity checks (USD monthly):

• Content program (regional): 4k - 15k
• Demand generation (mid-market, multi-channel): 8k - 30k
• Paid media (managed budget excl): fee 10% - 20% of ad spend, minimum 3k - 8k
• SEO (technical + content): 3k - 12k
• Full GTM program (strategy + ops + demand): 20k - 70k

Adjust for market complexity, language count, and required compliance work.

Key cost drivers and negotiation levers

Cost drivers:

• Number of markets and languages
• Need for local hires or field sales support
• Paid media spend and bidding complexity
• Required integrations and analytics setup

Negotiation levers:

• Commit to longer minimum terms in exchange for lower hourly rates
• Limit scope to core markets first
• Move certain tasks in-house after ramp
• Tie part of fee to milestones, not vague outcomes

When to use performance-based fees

Performance fees work when outputs are measurable and attributable. Use them for:

• MQLs that meet a strict, agreed definition
• New qualified logos
• Revenue or pipeline influenced by campaign, with clean tracking

Avoid performance fees if lead attribution is fuzzy or the sales cycle is long. If you do use them, cap the upside and set a baseline retainer so the agency can staff properly.

Go-to-Market Tactics for South America

Regional GTM is messy if you apply one-size-fits-all thinking.

Market prioritization criteria and mapping

Prioritize markets by these practical signals:

• Purchase power and buyer concentration
• Language and cultural proximity to existing assets
• Ease of doing business and local regulatory complexity
• Presence of channel partners or distribution networks

Map markets into tiers: 1 - test, 2 - scale, 3 - delay. Start with one Tier 1 market and one Tier 2 concurrently.

Localization checklist (language, messaging, payments)

• Language: translate and locally proofread. Spanish variants and Portuguese are not interchangeable.
• Messaging: adapt use cases to local buyers. Case study from a US enterprise might not resonate with a regional mid-market buyer.
• Payments: support local payment rails and tax invoices. If you ignore local billing preferences you lose deals.
• Legal: local contract templates and consumer laws can require tweaks.

Recommended channel mix by buyer stage

• Awareness: PR in local trade press, industry webinars, targeted LinkedIn content.
• Consideration: localized whitepapers, customer panels, solution briefings with local references.
• Decision: tailored demos, local trials, channel incentives.

Mix paid media for scale, owned content for credibility, and events for trust. Adjust spend by market maturity.

Regional talent, timezones, and partner use

Hire local for sales and customer-facing roles. Keep strategy and core content centralized until you validate messaging. Use partners for translations, local media buys, and tax/legal support. Be explicit about expectations for overlap hours; timezone mismatch is the killer of momentum.

Contracts, SLAs, and KPIs

Contracts should protect your business and keep the agency honest.

Must-have contract clauses (IP, data, exit)

• IP ownership: you must own creative and campaign assets from day one, or have an irrevocable license.
• Data access and portability: raw campaign and CRM data delivered regularly and structured.
• Exit terms: defined transition period, knowledge transfer obligations, and escrow of campaign assets.
• Confidentiality and non-solicit: limit poaching of your team for a set period.

Example SLA items (response, lead quality)

• Response time: initial email within 4 business hours, urgent issues within 1 hour.
• Lead delivery: all qualified leads delivered into CRM within 24 hours with source and tags.
• Lead quality: at least X% of delivered leads must meet agreed qualification criteria; below threshold triggers remediation plan.
• Campaign uptime: landing pages and tracking must be 99% available during campaign windows.

KPI framework and reporting cadence

Use a simple tiered KPI set:

• Activity KPIs: impressions, clicks, assets produced
• Funnel KPIs: MQLs, SQLs, demo rate
• Outcome KPIs: opportunities, influenced pipeline, ARR

Reporting cadence:

• Weekly: activity snapshot and blockers
• Monthly: performance review with trends and experiments
• Quarterly: strategic planning and budget reprioritization

Monthly review scorecard template

• Executive summary: one paragraph and three bullets
• Metrics table: target vs actual for top 5 KPIs
• RAG status per KPI with cause
• Top 3 wins and top 3 risks
• Action items owner and due date

Keep the review tight. If it’s an hour each month, you’ll actually do it.

Onboarding and Ramp Checklist

Onboarding is where accounts succeed or quietly fail.

30/60/90-day plan with milestones

30 days:

• Confirm SLAs and scorecard
• Full access to analytics, CRM, ad accounts
• First messaging and landing page live

60 days:

• Run first paid tests and two A/B landing experiments
• Establish lead qualification flow into sales
• Weekly stand-up rhythm locked

90 days:

• Deliver initial pipeline targets or a documented recovery plan
• Handoff internal playbooks for repeatable tactics
• Staffing adjustments if needed

Asset and access handover list

• CRM admin access and permission matrix
• Analytics and tag manager accounts
• Ad account admin or shared access
• Domain, DNS, and CDNs for landing pages
• Brand assets, legal templates, and approved copy library

Never let an agency control accounts exclusively. Shared access prevents hostage situations.

Early experiments and success metrics

Run small, fast tests:

• 2 creatives x 2 audiences for paid
• Headline A/B on top-performing page
• Short-form case study promoted to a narrow list

Success metrics: statistically meaningful lift in conversion or CPL improvement of at least 15% within test window.

Communication rhythm and governance roles

• Weekly tactical stand-up with delivery leads
• Monthly performance review with executives
• Quarter business review with strategic owners

Define who approves creative, budgets, and pause decisions. No ambiguous approvers.

Red Flags and Risk Mitigation

Watch behavior, not promises.

Operational warning signs to watch for

• High staff churn on your account
• Overreliance on founders for day-to-day work
• Refusal to provide raw data or access
• Constantly missed deadlines with new excuses

Legal, compliance, and data risks

• Cross-border data transfers without legal basis
• Lack of data processing agreement or subprocessor list
• Tax and invoicing issues in local jurisdictions

Have local counsel review contracts for sensitive markets.

Contingency planning and exit strategy

• Maintain backups of all creative, landing pages, and tracking tags
• Define knowledge transfer steps in contract with clear timelines
• Keep a parallel list of small vendors who can step in for critical services

A documented exit plan prevents a slow-motion train wreck.

How to validate references and claims

Don’t just call references. Do this:

• Ask for raw campaign reports showing funnel performance
• Request a short recorded walkthrough of their setup
• Speak to a client at the level you will interact with, not just procurement
• Check LinkedIn for employee tenure and team stability

If they refuse any of these, assume they are hiding something.