The Sneaker Proxy Playbook: ISP vs Residential for Nike, Shopify, and Footsites

Sneaker releases are time-compressed auctions. When a hyped Nike Dunk drops on SNKRS and sells out in 11 seconds, the proxy infrastructure behind your bot is not an optimization -- it is the entire margin between a successful checkout and an L. Every millisecond of latency, every blocked IP, every failed session costs real money on the resale market.

This guide is written for bot runners who want to understand the engineering behind proxy selection, not just which provider to buy. If you understand why ISP proxies outperform on Shopify but residential proxies survive longer on Footsites, you can adapt your setup to any platform change. And platforms change constantly.

For foundational knowledge on proxy types, see our ISP vs residential comparison. This playbook assumes you know the basics and focuses specifically on sneaker platform requirements.

The Sneaker Release Landscape in 2026

Before diving into proxy configuration, you need to understand how the major platforms handle bot traffic differently. Their detection strategies directly determine which proxy type works.

Nike SNKRS

Nike uses a custom anti-bot system built on top of Akamai Bot Manager. The system evaluates:

• Device fingerprinting: SNKRS collects detailed device telemetry including screen resolution, GPU renderer, installed fonts, and sensor data (on mobile). This fingerprint persists across IP changes.
• Behavioral scoring: The system tracks tap timing, scroll velocity, and navigation patterns. Automated interactions that lack natural variance get flagged.
• IP reputation: Nike maintains an IP reputation database that tracks IPs across drops. An IP that participated in a previous drop and showed bot-like behavior starts the next drop with a negative score.
• Queue randomization: SNKRS uses a lottery system (not FCFS) for most hyped releases, which means speed alone does not guarantee success. But getting into the queue requires a clean session, and that requires clean IPs.

Proxy implication: Nike's combination of device fingerprinting and IP reputation means that IP quality matters more than IP speed. Residential proxies have a higher success rate on SNKRS because their IPs have organic reputation histories.

Shopify (Kith, Bodega, Concepts, Notre, Undefeated)

Shopify's bot protection comes from Shopify's built-in "bot protection" feature plus the individual store's Cloudflare or Kasada configuration. The critical technical detail:

• Shopify checkout is FCFS. Unlike SNKRS' lottery, Shopify drops are first-come-first-served. Speed is the decisive factor.
• Checkout flow requires session persistence. Adding to cart, entering shipping info, and processing payment are a multi-step flow. The IP must remain consistent throughout (typically 30-60 seconds).
• Rate limiting per IP. Shopify throttles requests per IP. Each IP can handle a limited number of add-to-cart attempts per second.

Proxy implication: Shopify rewards speed and session consistency. ISP proxies with their 50-250ms response times and static IP assignments are the clear winner. You need sticky sessions that maintain the same IP through the entire checkout flow.

Footsites (Foot Locker, Champs, Eastbay)

Footsites use Akamai Bot Manager with aggressive fingerprinting:

• Sensor data collection: Footsites collect extensive JavaScript sensor data (similar to Nike).
• IP subnet blocking: Footsites are known to block entire /24 subnets when they detect bot activity from multiple IPs in the same range. This makes subnet diversity critical.
• Queue system: Footsites use a virtual waiting room for hyped releases. Entry into the queue requires passing an Akamai challenge.

Proxy implication: Footsites' subnet-level blocking means you need IPs from diverse ASNs and subnets. Residential proxies' inherent diversity (each IP from a different household, different ISP) provides natural protection against subnet bans. ISP proxies can work but only if sourced from multiple providers with distinct subnets.

Yeezy Supply / Adidas Confirmed

Adidas uses a combination of Akamai and custom anti-bot with a queue system:

• Queue fairness checks: The system periodically re-verifies sessions in the queue. An IP change during the queue wait can result in ejection.
• Geographic restrictions: Some releases are region-locked. The proxy IP must match the release region.
• Session duration: Queue waits can last 30+ minutes. Sticky sessions must hold for the entire duration.

Proxy implication: Long session requirements favor ISP proxies with permanent static IPs over residential sticky sessions that may time out. Geographic targeting is mandatory.

ISP vs Residential: Platform-by-Platform Recommendation

Platform	Recommended Proxy	Reason	Session Type
Nike SNKRS	Residential	IP reputation matters most; lottery system reduces speed advantage	Per-request or short sticky (2-5 min)
Shopify stores	ISP	Speed is decisive in FCFS; checkout needs consistent IP	Sticky (static IP)
Foot Locker / Champs	Residential	Subnet blocking; needs IP diversity	Short sticky (5-10 min)
Yeezy Supply / Confirmed	ISP	Long queue waits need permanent IP; geographic lock	Static IP with geo-targeting
Amazon (limited drops)	Residential	Advanced anti-bot (DataDome); high IP reputation requirements	Per-request rotation

Proxy Configuration for Sneaker Bots

ISP Proxy Setup for Shopify

For Shopify drops, you want one dedicated ISP IP per bot task. Each task runs an independent checkout flow on a persistent IP.

import requests

# Each task gets its own static ISP proxy IP
# With ISP proxies, each IP in your plan is dedicated to you
TASK_PROXIES = {
    "task_1": "http://USER:PASS@gate.hexproxies.com:8080",
    "task_2": "http://USER:PASS@gate.hexproxies.com:8080",
    "task_3": "http://USER:PASS@gate.hexproxies.com:8080",
}

class ShopifyCheckout:
    def __init__(self, task_id: str, proxy_url: str):
        self.session = requests.Session()
        self.session.proxies = {
            "http": proxy_url,
            "https": proxy_url,
        }
        self.session.headers.update({
            "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) "
                          "AppleWebKit/537.36 (KHTML, like Gecko) "
                          "Chrome/124.0.0.0 Safari/537.36",
            "Accept": "application/json",
        })
    
    def add_to_cart(self, variant_id: str, store_url: str) -> bool:
        """Add item to cart. Speed-critical -- ISP latency advantage matters here."""
        payload = {"id": variant_id, "quantity": 1}
        response = self.session.post(
            f"{store_url}/cart/add.js",
            json=payload,
            timeout=5,
        )
        return response.status_code == 200
    
    def begin_checkout(self, store_url: str) -> str:
        """Create checkout -- must use same IP as add_to_cart."""
        response = self.session.post(
            f"{store_url}/checkout",
            timeout=10,
        )
        return response.url  # Returns checkout URL with token

The critical detail: the requests.Session() object maintains cookies across requests, and the ISP proxy maintains the same IP. Both layers of persistence are necessary for Shopify checkouts.

Residential Proxy Setup for Nike SNKRS

For SNKRS, each entry needs a unique residential IP with a clean reputation. Per-request rotation gives each entry a fresh IP from the residential pool.

import requests
import uuid

def create_snkrs_session() -> requests.Session:
    """Each SNKRS entry gets a unique session with a unique residential IP."""
    session_id = uuid.uuid4().hex[:12]
    
    session = requests.Session()
    session.proxies = {
        "http": f"http://USER-session-{session_id}-country-us:PASS@gate.hexproxies.com:8080",
        "https": f"http://USER-session-{session_id}-country-us:PASS@gate.hexproxies.com:8080",
    }
    session.headers.update({
        "User-Agent": "Mozilla/5.0 (iPhone; CPU iPhone OS 17_4 like Mac OS X) "
                      "AppleWebKit/605.1.15 (KHTML, like Gecko) "
                      "Version/17.4 Mobile/15E148 Safari/604.1",
    })
    return session

Note the country-us parameter. SNKRS validates that the IP's geographic location matches the account's shipping address region. A US account entering a draw from a German IP will be flagged.

Residential Proxy Setup for Footsites

Footsites require the most careful proxy management due to subnet blocking. The key is ensuring maximum IP diversity.

import uuid

def create_footsite_sessions(count: int) -> list:
    """
    Create multiple sessions with maximum IP diversity.
    Each session gets a unique session ID, ensuring a unique IP
    from the residential pool. The pool's natural subnet diversity
    provides protection against /24 blocks.
    """
    sessions = []
    for _ in range(count):
        session_id = uuid.uuid4().hex[:12]
        session = requests.Session()
        session.proxies = {
            "http": f"http://USER-session-{session_id}-country-us:PASS@gate.hexproxies.com:8080",
            "https": f"http://USER-session-{session_id}-country-us:PASS@gate.hexproxies.com:8080",
        }
        sessions.append(session)
    return sessions

With residential proxies, each session ID maps to a different IP from the pool. Since residential IPs are sourced from millions of real consumer connections across thousands of ISPs, the probability of two sessions landing on the same /24 subnet is negligible. This is a structural advantage that ISP proxies cannot easily replicate because ISP proxy pools are smaller and often concentrated in fewer subnets.

Latency Benchmarks: Why Milliseconds Matter on Shopify

We benchmarked proxy response times against a Shopify /cart/add.js endpoint (the most latency-sensitive request in a sneaker bot flow):

Proxy Type	Median Latency	P95 Latency	P99 Latency
ISP (US-based)	87ms	210ms	380ms
Residential (US-based)	340ms	890ms	1,420ms
No proxy (direct)	45ms	120ms	250ms

Source: Hex Proxies internal testing, March 2026. 10,000 requests per proxy type.

The 253ms median difference between ISP and residential proxies does not sound like much. But in a FCFS Shopify drop with 50,000 bots competing for 500 pairs, that 253ms is the difference between being in the first 1,000 checkout attempts and being in the 10,000th. Shopify processes checkouts sequentially per variant, so queue position is everything.

The Speed vs Reputation Trade-off

This is the fundamental tension in sneaker proxy selection:

• ISP proxies give you speed (87ms median) but lower bypass rates on advanced anti-bot systems (72-88% success against Akamai-protected sites).
• Residential proxies give you reputation (91-95% success against Akamai) but at the cost of 3-4x higher latency.

The resolution is platform-specific. On Shopify (FCFS), speed wins because you need to beat other bots to the checkout server. On Nike SNKRS (lottery), reputation wins because getting into the draw is the bottleneck, not speed.

Session Management Strategies

One-IP-Per-Task Model (Shopify, Yeezy Supply)

Each bot task (one size, one checkout profile) gets a dedicated static IP. The task uses a persistent session with cookies and the same IP throughout.

When to use: FCFS platforms where checkout requires IP consistency.

Proxy type: ISP proxies. One IP per task means you need as many ISP IPs as concurrent tasks. Running 50 tasks = 50 ISP IPs.

Cost at Hex Proxies: 50 ISP IPs at $2.08-$2.47/IP = $104-$123.50/month. Unlimited bandwidth means aggressive retries during drops cost nothing extra.

One-IP-Per-Entry Model (SNKRS, Footsites)

Each raffle entry gets a unique residential IP via a unique session ID. The IP persists for the duration of the entry process (typically 2-5 minutes).

When to use: Lottery-based platforms where each entry should appear to come from a different person.

Proxy type: Residential proxies with sticky sessions. Session duration should be set to 10 minutes (5-minute buffer beyond the entry process).

Cost at Hex Proxies: 100 entries at approximately 5 MB per entry = 500 MB. At $4.25-$4.75/GB, that is roughly $2.13-$2.38 per drop. Minimal cost because each entry transfers very little data.

Rotating IP Model (Monitoring and Research)

For monitoring restock pages, checking product availability, and scraping release calendars, use per-request rotation with no session persistence.

When to use: Pre-drop research and restock monitoring where you do not need session continuity.

Proxy type: Residential proxies with default rotation. Each request gets a fresh IP, minimizing the risk of any single IP being flagged.

Ban Detection and Recovery

Even with optimal proxy configuration, bans happen. The difference between experienced and novice bot runners is how quickly they detect and recover from bans.

Detecting Bans Before They Cost You

HTTP 403 responses. The obvious signal. If your success rate drops below 90% on a target, investigate immediately.

Soft bans (delayed responses). Some platforms do not return 403. Instead, they slow down responses to bot-like traffic. If your median latency suddenly doubles on a specific platform, you may be soft-banned.

CAPTCHA frequency increase. Getting one CAPTCHA per 100 requests is normal. Getting one per 5 requests means the platform has flagged your IP pattern.

Queue position regression. On platforms with virtual queues, bot-flagged sessions get placed at the back of the queue or in a "bot queue" that never advances.

Recovery Tactics

For ISP proxies: If an ISP IP is banned, it stays banned because it is static. You need to rotate to a different ISP IP from your pool, or request a replacement IP from your provider. This is why sneaker runners should maintain 20-30% more ISP IPs than active tasks -- the extras are your reserve bench.

For residential proxies: A banned residential IP is automatically replaced on the next request (with per-request rotation) or when the sticky session expires. The massive pool size means the chance of receiving the same banned IP again is near zero.

Subnet rotation for footsites: If you suspect a /24 subnet ban on a footsite, you need IPs from entirely different subnets. With residential proxies, request IPs from a different geographic region (different state or city) to ensure different ISP infrastructure and subnets.

Cost Analysis: Real Sneaker Bot Scenarios

Scenario 1: Shopify-Focused Runner (20 drops/month)

• 30 bot tasks per drop
• ISP proxies needed: 30 IPs (plus 10 reserve) = 40 ISP IPs
• Monthly cost: 40 IPs x $2.08-$2.47 = $83.20-$98.80/month
• Cost per drop: $4.16-$4.94

Scenario 2: SNKRS-Focused Runner (8 drops/month)

• 50 entries per drop
• Residential bandwidth per drop: ~250 MB
• Monthly bandwidth: 2 GB
• Monthly cost: 2 GB x $4.25-$4.75 = $8.50-$9.50/month
• Cost per drop: $1.06-$1.19

Scenario 3: Multi-Platform Runner (All platforms)

• 40 ISP IPs for Shopify + Yeezy: $83.20-$98.80/month
• 5 GB residential for SNKRS + Footsites: $21.25-$23.75/month
• Total: $104.45-$122.55/month

At these costs, a single successful Shopify checkout on a hyped release (typical resale profit: $100-$500 per pair) covers months of proxy expenses.

Platform-Specific Tips

Shopify Quick Tips

1. Pre-load cookies by visiting the site 30-60 minutes before the drop through your ISP proxy. This builds session history.
2. Use the fastest available ISP region. If the Shopify store's servers are in US-East, use US-East ISP proxies.
3. Monitor /products.json for variant IDs before the drop. This endpoint is less aggressively rate-limited.

SNKRS Quick Tips

1. Warm up accounts on the SNKRS app days before the drop. Browse products, save favorites, check sizes. Do this through your proxy to build IP-account association.
2. Use mobile User-Agents. SNKRS heavily favors mobile app traffic.
3. Match proxy country to account country precisely. US account = US residential IP.

Footsite Quick Tips

1. Avoid ISP proxies on Footsites unless you have verified the specific IP subnets are not on their blocklist.
2. Generate new session IDs for every drop. Reusing session IDs from a previous drop risks inheriting reputation from that drop's activity.
3. Use residential proxies from Tier 1 US ISPs when targeting US Footsites. The IPs are less likely to be flagged.

Frequently Asked Questions

How many ISP IPs do I need per Shopify task?

One IP per task. If you run 30 concurrent tasks, you need 30 ISP IPs minimum. We recommend 30-50% extra as reserves for ban recovery. See our ISP proxy plans for volume pricing.

Can I use the same proxies for SNKRS and Shopify?

You can, but you should not. SNKRS and Shopify have different optimal proxy types. Use residential for SNKRS and ISP for Shopify. The cost of maintaining both is minimal compared to the checkout success rate improvement.

Do proxy location and sneaker region need to match?

Yes, for SNKRS and Yeezy Supply. These platforms validate that the IP's geographic location is consistent with the account's shipping address. For Shopify, geographic matching is less critical but still recommended for minimizing latency.

What happens if my ISP IP gets banned during a drop?

Your bot should have a fallback IP configured. Most sneaker bots support proxy lists -- put your reserve IPs in the fallback positions. The task will fail on the banned IP and can be retried on a fresh IP. This is why maintaining reserve IPs is essential.

---

Ready to set up your sneaker proxy stack? Start with ISP proxies for Shopify speed at $2.08-$2.47/IP, or residential proxies at $4.25-$4.75/GB for SNKRS and Footsite diversity. Check our sneaker bot proxy guide for additional platform-specific configurations.