Cybersecurity credentials
NCA registration
Registered cybersecurity service provider

ISO/IEC 27001
ISO/IEC 27001 certified
The exposure
The doors into an online store that merchants forget they installed
Each of these ships with the platform or arrives with growth, and each reaches money or customer data. The control column is where they are caught — when someone owns it.
Five exposure routes — admin panel, plugins, third-party scripts, customer accounts, bot traffic — converging on the store's control points, producing genuine orders, an uncompromised checkout and uptime through campaigns.
The admin panel
One login away from prices, orders and exports — often on a guessable URL with no second factor.
Plugins and themes
Code from dozens of authors running with the store's full trust, patched on their schedule, not yours.
Third-party scripts
Analytics, chat and pixels loaded onto the checkout, each answering to a server you do not control.
Customer accounts
Saved cards, addresses and loyalty balances behind passwords reused from breached services.
Bot traffic
Stuffing, scraping and scalping arriving as ordinary requests, indistinguishable from a good day.
What we do
The store's control points
Hardened admin access, a plugin inventory with an update owner, a script policy on the payment path, abuse limits on accounts, and bot defences tuned to your catalogue.
Orders that are genuine
Stock sold to customers, not to scripts.
A checkout that keeps its secrets
Card data that reaches the processor and nobody else.
A store that stays up mid-campaign
Peak traffic absorbed instead of exploited.
The payment page
Every script on your checkout answers to whoever controls its server
A modern checkout page loads code from many places: the platform, an analytics vendor, a chat widget, a marketing pixel, a reviews service. Each of those scripts runs with full access to the page — including the form where a customer types a card number. Digital skimming works exactly there: compromise one script vendor, or one forgotten tag, and every card typed into the page is quietly copied out while the payment itself succeeds. Nothing breaks. The merchant finds out months later, from the card schemes, as the common point of purchase in a fraud pattern — which is the most expensive possible way to learn what was on your own page.
The defence is to treat the payment page as a controlled zone. Inventory what loads there and strike off everything that does not need to be on that one page — the marketing pixel can live on the confirmation page instead. Pin what remains: integrity attributes and a content security policy so a changed or unauthorised script fails loudly instead of running silently. Then watch it, because the page changes with every campaign tag the marketing team adds. A checkout that loads five known scripts under a policy that screams on the sixth is a fundamentally different risk from one that loads whatever the tag manager was told to load this week.
Abuse as traffic
Stuffing, scalping and loyalty drains: the attacks that arrive as customers
These are economic attacks: each one profits from your store working exactly as designed. The defence is economic too — raising the attacker's cost per attempt until your store stops being the cheap target.
Credential stuffing against customer logins
Password lists from other breaches, replayed against your login at machine speed. Each hit yields saved cards, addresses and loyalty balances. Rate limits by source and identity, breached-password checks and a second factor for stored-value actions turn a bulk business into a retail one.
Scalping bots that buy the drop before people can
Scripts that add to cart and check out in the first seconds of a launch, emptying stock for resale. The customers you ran the campaign for meet a sold-out page and remember it. Queue fairness, purchase limits enforced server-side and bot challenges on the cart path keep the drop for humans.
Loyalty and gift-card balances spent by strangers
Points and gift cards are money with weaker locks: guessable codes, no second factor, alerts nobody reads. Attackers drain them quietly because customers check balances rarely. Treat stored value like payment value — strong codes, redemption limits, notifications on spend — and the drain stops paying.
Refund and promotion abuse at the margins
Promo codes farmed across fake accounts, refund claims for items that arrived, item-not-received played against every courier gap. Individually small, collectively a margin line. The fix is correlation — devices, addresses and payment instruments linked across accounts — plus limits that tighten with anomaly.
When it happens
An incident plan priced in minutes of lost checkout
Retail incident response has a clock other industries do not: every minute of a broken or untrusted checkout is revenue that does not come back, and the incident will choose the worst minute — mid-campaign, Ramadan, White Friday, launch night. That deadline changes the plan's shape. The decisions that take longest in a crisis are not technical; they are permissions. Who may take the checkout offline? Who may disable a plugin that fulfilment depends on? At what evidence threshold do we notify the payment provider, and who speaks to customers? A plan that answers those in advance — with names, not committees — converts hours of escalation into minutes of action.
The second retail-specific discipline is the degraded mode. A store rarely has to choose between fully up and fully down: card capture can switch to the processor's hosted page while your own checkout is suspect, a compromised plugin can be disabled while its feature is stubbed, sales can continue with loyalty redemption frozen while balances are verified. Each of those degraded modes has to be designed and rehearsed in advance, because inventing one during the incident is how a two-hour problem becomes a two-day outage. The stores that come through an incident with their trading intact are not the ones that were never hit; they are the ones that had already decided what to switch off.
Our role
What we harden across a store and which trade-offs stay commercial
- 01
The whole machine, inventoried
We assess the store as an attacker maps it — admin surfaces, plugin estate, payment-page scripts, account and loyalty flows, bot exposure — and rank findings by what each would cost per hour of a live campaign.
- 02
Controls that respect conversion
Every defence is designed against the funnel it sits in: bot challenges that genuine customers rarely meet, step-up placed on stored value rather than on browsing, a script policy that blocks skimmers without breaking the tags marketing needs.
- 03
A rehearsed bad day
We build and exercise the incident plan in retail time — named authorities, designed degraded modes, provider contacts ready — so the first campaign-night incident is executed from a script rather than improvised.
How much friction the funnel tolerates, and which campaigns justify which risks, are trading decisions your commercial team owns; we make sure they are taken knowing what each door left open would cost.
Common questions
What buyers ask
How would we know if a skimmer was running on our checkout right now?+
Without controls, you mostly would not — that is what makes skimming profitable. The payment succeeds, the customer is happy, and the copy of the card leaves in the background; discovery usually comes months later from the card schemes tracing fraud back to a common point of purchase. The immediate check is an inventory: load your checkout as a customer and list every script that executes, then account for each one — who owns it, why it is on that page, whether it has changed recently. The durable answer is a content security policy and integrity pinning on the payment path, plus monitoring that alerts when the set of scripts changes. At that point a skimmer stops being silent: the unauthorised script fails to load, and you hear about it the day it is attempted rather than the quarter after it worked.
Bots keep buying out our launches — is that a security problem or a marketing one?+
Both, which is why it usually goes unsolved: marketing sees sold-out numbers that look like success, and security sees traffic that is not an intrusion, so nobody owns it. Treat it as an economics problem with a security toolkit. The scalper profits when the cost of a thousand automated checkouts is lower than the resale margin; the defence raises that cost — queue fairness that randomises rather than rewards speed, purchase limits enforced server-side where a script cannot lie about them, challenges on the cart path tuned to bot signals, and correlation across the accounts, devices and payment instruments a farm reuses. The measure of success is not zero bots; it is launches where the stock ends up with the customers the campaign was priced for, which is a revenue outcome marketing can see.
We run on a hosted commerce platform — what part of security is still ours?+
More than the brochure suggests. The platform secures its infrastructure: servers, patching of the core, the availability of the service. What it cannot secure is how you use it — and that is where stores are actually breached. Your admin accounts and their second factor, the staff and agency access that never gets revoked, every plugin and theme you install, every script your tag manager injects onto the checkout, the API keys handed to integrations, and the abuse of your accounts and promotions are all yours regardless of who hosts the core. A useful exercise is to write the split down: for each part of the store, who patches it, who watches it, who acts when it misbehaves. The rows with your name on them are usually more numerous than expected, and the rows with nobody's name on them are the findings.
In this cluster
Map your store's exposure
An exposure review of your store that inventories the admin surfaces, plugins and payment-page scripts an attacker would try first, ranked by what each one could cost per hour of a campaign.
What happens next
- 1We reply within one business day.
- 2A 30-minute scoping call with a senior consultant.
- 3A written proposal with scope, team, and timeline.
Or reach us directly
