Claude-code-plugins klaviyo-cost-tuning

install
source · Clone the upstream repo
git clone https://github.com/jeremylongshore/claude-code-plugins-plus-skills
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/jeremylongshore/claude-code-plugins-plus-skills "$T" && mkdir -p ~/.claude/skills && cp -r "$T/plugins/saas-packs/klaviyo-pack/skills/klaviyo-cost-tuning" ~/.claude/skills/jeremylongshore-claude-code-plugins-klaviyo-cost-tuning && rm -rf "$T"
manifest: plugins/saas-packs/klaviyo-pack/skills/klaviyo-cost-tuning/SKILL.md
source content

Klaviyo Cost Tuning

Overview

Optimize Klaviyo costs through active profile management, list hygiene, event sampling, and API usage monitoring. Klaviyo bills primarily by active profiles and message volume, not API calls.

Prerequisites

  • Access to Klaviyo billing dashboard
  • Understanding of active profile definition
  • klaviyo-api
    SDK for programmatic management

Klaviyo Pricing Model

Klaviyo bills based on active profiles (contacts who have received or been targeted by marketing), not API requests.

ComponentHow It's BilledCost Driver
EmailPer active profile tierNumber of marketable profiles
SMSPer message sent + carrier feesMessage volume
PushIncluded with email planN/A
API callsFree (rate limited, not billed)N/A
ReviewsPer request volumeReview request sends

Email Pricing Tiers (Approximate)

Active ProfilesMonthly Cost
0 - 250Free
251 - 500$20/mo
501 - 1,000$30/mo
1,001 - 1,500$45/mo
1,501 - 5,000$60-$100/mo
5,001 - 10,000$100-$150/mo
10,001 - 25,000$150-$375/mo
25,001+Custom pricing

Key insight: Reducing active profiles has the biggest cost impact. Cleaning suppressed/unengaged contacts directly reduces your bill.

Instructions

Step 1: Audit Active Profile Count

import { ApiKeySession, ProfilesApi, SegmentsApi } from 'klaviyo-api';

const session = new ApiKeySession(process.env.KLAVIYO_PRIVATE_KEY!);
const profilesApi = new ProfilesApi(session);

// Count total profiles
let totalProfiles = 0;
let cursor: string | undefined;
do {
  const response = await profilesApi.getProfiles({
    pageCursor: cursor,
    fieldsProfile: ['email'],  // Minimal fields for speed
  });
  totalProfiles += response.body.data.length;
  const nextLink = response.body.links?.next;
  cursor = nextLink ? new URL(nextLink).searchParams.get('page[cursor]') || undefined : undefined;
} while (cursor);

console.log(`Total profiles: ${totalProfiles}`);

Step 2: Identify Unengaged Profiles

// Find profiles that haven't opened/clicked in 180+ days
// Create a segment in Klaviyo for this, then query it
const segmentsApi = new SegmentsApi(session);

const segments = await segmentsApi.getSegments({
  filter: 'equals(name,"Unengaged 180+ Days")',
});

if (segments.body.data.length > 0) {
  const segmentId = segments.body.data[0].id;
  const unengaged = await segmentsApi.getSegmentProfiles({
    id: segmentId,
    fieldsProfile: ['email', 'created'],
  });
  console.log(`Unengaged profiles: ${unengaged.body.data.length}+`);
}

Step 3: Suppress Unengaged Contacts

// Move unengaged profiles to a suppressed list (removes from active count)
import { ListsApi, ListEnum, ProfileEnum } from 'klaviyo-api';

const listsApi = new ListsApi(session);

// Option 1: Unsubscribe (profile stays but isn't marketable = not billed)
await profilesApi.unsubscribeProfiles({
  data: {
    type: 'profile-subscription-bulk-delete-job',
    attributes: {
      profiles: {
        data: unengagedEmails.map(email => ({
          type: ProfileEnum.Profile,
          attributes: {
            email,
            subscriptions: {
              email: { marketing: { consent: 'UNSUBSCRIBED' } },
            },
          },
        })),
      },
    },
    relationships: {
      list: { data: { type: ListEnum.List, id: 'MAIN_LIST_ID' } },
    },
  },
});

// Option 2: Suppress via profile update (add to global suppression)
for (const email of unengagedEmails) {
  await profilesApi.createOrUpdateProfile({
    data: {
      type: ProfileEnum.Profile,
      attributes: {
        email,
        properties: { suppressedAt: new Date().toISOString(), suppressReason: 'unengaged-180d' },
      },
    },
  });
}

Step 4: Event Sampling for Non-Critical Tracking

// Not all events need to be tracked -- sample non-critical ones
function shouldTrackEvent(eventName: string, samplingRates: Record<string, number>): boolean {
  const rate = samplingRates[eventName] ?? 1.0;  // Default: track everything
  return Math.random() < rate;
}

const samplingConfig = {
  'Placed Order': 1.0,       // Always track (revenue attribution)
  'Started Checkout': 1.0,   // Always track (cart abandonment)
  'Viewed Product': 0.25,    // 25% sample (high volume, less critical)
  'Page View': 0.1,          // 10% sample (very high volume)
};

// Before tracking
if (shouldTrackEvent('Viewed Product', samplingConfig)) {
  await eventsApi.createEvent({ /* ... */ });
}

Step 5: API Usage Monitor

// Track API call volume to detect runaway processes
class KlaviyoUsageTracker {
  private callCount = 0;
  private readonly startTime = Date.now();

  track(): void {
    this.callCount++;
    // Warn if approaching steady rate limit
    const elapsedMinutes = (Date.now() - this.startTime) / 60000;
    const ratePerMinute = this.callCount / Math.max(elapsedMinutes, 1);

    if (ratePerMinute > 500) {
      console.warn(`[Klaviyo] High API rate: ${Math.round(ratePerMinute)} req/min (limit: 700)`);
    }
  }

  getStats(): { totalCalls: number; ratePerMinute: number } {
    const elapsedMinutes = (Date.now() - this.startTime) / 60000;
    return {
      totalCalls: this.callCount,
      ratePerMinute: Math.round(this.callCount / Math.max(elapsedMinutes, 1)),
    };
  }
}

export const usageTracker = new KlaviyoUsageTracker();

Cost Reduction Checklist

  • Suppress profiles unengaged >180 days
  • Remove hard-bounced email addresses
  • Audit and merge duplicate profiles
  • Use double opt-in to reduce fake signups
  • Sample high-volume, low-value events
  • Batch API calls instead of individual requests
  • Cache frequently-read data (segments, lists)
  • Use sparse fieldsets to reduce transfer size
  • Review SMS sending -- highest per-message cost
  • Set up sunset flow (auto-suppress after N days unengaged)

Error Handling

IssueCauseSolution
Unexpected bill increaseUnengaged profiles grewRun suppression script
SMS costs spikingFlow sending to full listAdd engaged-only segment filter
Duplicate profilesMultiple identify callsMerge duplicates, use
createOrUpdateProfile
API rate limits hitBulk operationsUse queue with concurrency control

Resources

Next Steps

For architecture patterns, see

klaviyo-reference-architecture
.