Rate Limits

Overview

The Trakkr API limits how many requests it accepts per minute, so one busy caller cannot slow the service down for everyone else.

How the limit works

The main limit is a per-minute count on each endpoint, listed in the table below. It is the same on every plan. There is no burst allowance and no higher tier that lifts these numbers. API access requires the Scale plan ($500/mo) or higher, but the plan does not change the per-minute figures.

Limits are counted per account. Every request made with your API key shares one budget, whichever servers or IP addresses it comes from, and other customers on the same network never use yours.

Creating brands has one more limit, shared with the app: each account can add up to twice its brand slots in new brands every 30 days, or its slots plus five on smaller plans. Brands you later delete still count, and restoring a deleted brand does not. A create past the limit returns 429 with brand_creation_limit_reached and the date the next one fits. See Brands and client portals.

Per-Endpoint Limits

Different endpoints have different rate limits based on their computational cost. Write operations and resource-intensive endpoints have stricter limits.

Handling Rate Limits

When you go over, the API returns 429 Too Many Requests. The body holds one error string naming the limit you hit, for example Rate limit exceeded: 60 per 1 minute. Note that this is the one error that does not use the usual detail field.

Responses carry no rate limit headers. There is no X-RateLimit-Remaining, X-RateLimit-Limit, X-RateLimit-Reset or Retry-After to read, on any response including the 429. Plan around the published limits instead of measuring what is left.

Exponential Backoff

Exponential backoff means each retry waits longer than the last. Start above the one-minute window, then double, and add random jitter so clients do not all retry at exactly the same time:

# Implement exponential backoff in shell
attempt=1
max_attempts=5

while [ $attempt -le $max_attempts ]; do
  response=$(curl -s -w "%{http_code}" \
    -H "Authorization: Bearer $TRAKKR_API_KEY" \
    'https://api.trakkr.ai/get-brands')
  
  status=$(echo "$response" | tail -c 4)
  
  if [ "$status" -eq 200 ]; then
    echo "Success!"
    break
  elif [ "$status" -eq 429 ]; then
    sleep_time=$((2 ** $attempt))
    echo "Rate limited. Retrying in ${sleep_time}s..."
    sleep $sleep_time
    attempt=$((attempt + 1))
  else
    echo "Error: $status"
    break
  fi
done

Best Practices