Alex is Sprintlaw's co-founder and principal lawyer. Alex previously worked at a top-tier firm as a lawyer specialising in technology and media contracts, and founded a digital agency which he sold in 2015.
If your UK business collects, uses, or analyses data about people-whether they’re staff, customers, or website visitors-you’ve probably heard about GDPR. But you might not know that “profiling a business” (a term used under GDPR for automated analysis) can trigger extra compliance duties you shouldn’t ignore.
Maybe you run an online store that tracks buying habits, a gym that monitors member attendance, or a consultancy looking to launch targeted marketing. In all these cases, profiling can add significant legal responsibilities to your plate-especially around transparency, consent, and individual rights.
Getting your legal foundations right when it comes to profiling can feel tricky. But with some clear steps, you can stay compliant and build trust with your clients and team. Let’s break down what “profiling a business” really means, when you need to follow the rules, and how to protect both your business and your customers’ data from day one.
- What Is Profiling a Business Under GDPR?
- Do You Need Consent To Profile Customers Or Business Contacts?
- Can You Use Public Professional Profiles, Companies House Data Or LinkedIn-Style Data?
- When Does Profiling Become Automated Decision-Making?
- What Documents And Contracts Do You Need For Profiling?
- Key Takeaways
What Is Profiling a Business Under GDPR?
Under UK GDPR, profiling means automated processing of personal data to evaluate personal aspects of a natural person. That can include analysing or predicting a person's work performance, economic situation, health, preferences, interests, reliability, behaviour, location or movements.
In plain English, profiling is when software helps you build a picture of a person or put them into a category. It doesn't need to involve artificial intelligence or a complex algorithm. Everyday small business tools can involve profiling, including:
- Segmenting a mailing list by purchase history, location or browsing behaviour
- Using website analytics, cookies or pixels to understand visitor behaviour
- Creating lead scores in a CRM based on job title, company size or engagement
- Running credit checks, fraud checks or risk ratings
- Using recruitment software to rank or filter job applicants
- Monitoring employee performance, productivity or attendance through software
- Recommending products, discounts or content based on customer behaviour
UK GDPR still applies even if the tool is supplied by a third party. If the profiling is done for your business, or you decide why and how the data is used, you'll usually be a controller for at least part of that activity. If a marketing agency, analytics platform or AI tool processes personal data for you, you also need to check your processor contract and due diligence.
Profiling a Company Versus Profiling an Individual: What Is the Difference?
UK GDPR protects living individuals. It doesn't protect a limited company as a person in its own right. This means researching a company's turnover, sector, office locations or Companies House filings is not, by itself, UK GDPR profiling.
However, business data often includes personal data. For example, a director's name on Companies House, a founder's LinkedIn-style profile, a named procurement manager's email address, a sole trader's business details and an employee's job history can all be personal data if they identify a living person.
That distinction matters in practice. You can build a corporate profile for sales, risk assessment or market research, but the moment you attach views, scores or predictions to identifiable people, UK GDPR is in play.
What About Racial Profiling And Sensitive Data?
Racial profiling isn't just a reputational risk. Racial or ethnic origin is special category data under UK GDPR, which receives extra protection. Profiling people by racial or ethnic origin, health, biometric data used to uniquely identify them, religion, political opinions, trade union membership, sex life or sexual orientation is high risk and needs a separate Article 9 condition, such as explicit consent or another specific Article 9 condition.
For employers, profiling that risks discriminatory treatment can also create employment law issues. If a tool filters applicants, scores employees or targets customers in a way that disadvantages people because of protected characteristics, you should stop and get the setup reviewed before using the results.
Do You Need Consent To Profile Customers Or Business Contacts?
Not always. Consent is one lawful basis, but many businesses use legitimate interests for lower-risk profiling, such as B2B lead scoring, customer segmentation or measuring website engagement. The key is that your interest must be real, the profiling must be necessary for that purpose, and your interests must not be overridden by the person's rights and expectations.
You're more likely to need consent where:
- You use non-essential cookies, tracking pixels or similar technologies. PECR consent is usually required for these, separate from UK GDPR, although since 5 February 2026 some analytics cookies are exempt if you give clear information and a simple way to object.
- You send unsolicited electronic mail marketing to individual subscribers, unless a specific exception such as the soft opt-in applies.
- You use special category data, such as health or ethnicity, for profiling.
- You make high-impact decisions based on profiling.
- Your profiling would surprise people or feel intrusive in the context.
For direct marketing, UK GDPR gives people a strong right to object. If someone objects to direct marketing, you must stop processing their personal data for that purpose, including related profiling. This isn't a balancing exercise. Once the objection is made, that marketing profiling must stop for that person.
If you rely on consent, it needs to be specific, informed, freely given and easy to withdraw. Pre-ticked boxes, bundled consent or forcing someone to agree to profiling that isn't genuinely needed won't usually work. For more, check out our step-by-step guide to GDPR consent.
Can You Use Public Professional Profiles, Companies House Data Or LinkedIn-Style Data?
Publicly available doesn't mean free to use for any purpose. You may be able to use public professional profiles or Companies House information for B2B lead generation, recruitment research or supplier due diligence, but you still need a lawful basis, you still need to be fair, and you still need to tell people what you're doing where UK GDPR requires it.
For example, if you collect a named operations manager's work email from a conference website and add them to a CRM score for outbound sales, you're processing personal data. If you then infer that they're a high-value prospect based on job title, sector and behaviour, you're likely profiling them.
A practical way to handle this is to make your first contact short, relevant and transparent. Tell the person who you are, where you got their details or the category of source, why you're contacting them, how they can opt out and where to find your Privacy Policy. Also check the platform terms. Some professional networking sites restrict scraping, automated collection or use outside the platform.
If you're planning to contact someone about a professional profile, start by asking whether consent is actually the right basis. Viewing a public business profile may not require consent under UK GDPR, but collecting, storing, scoring or marketing to that person does need a lawful basis and fair notice.
When Does Profiling Become Automated Decision-Making?
Profiling becomes more legally sensitive when it feeds into a significant decision about a person, especially where there's no meaningful human involvement. Since 5 February 2026, the old UK GDPR Article 22 wording has been replaced by Articles 22A to 22D through the Data (Use and Access) Act 2025.
A significant decision is one that produces a legal effect or a similarly significant effect for the person. Examples can include refusing credit, rejecting a job applicant, setting insurance eligibility, terminating access to an essential service or making a major employment decision.
A decision is solely automated if there's no meaningful human involvement. A person rubber-stamping a software recommendation is unlikely to be enough. The human reviewer needs enough authority, information and time to genuinely assess the recommendation and change the outcome.
If your business makes a significant decision based solely on automated processing, you need safeguards. These include giving the person information, allowing them to make representations, enabling human intervention and allowing them to contest the decision. If the decision is based entirely or partly on special category data, stricter conditions apply, such as explicit consent or a permitted legal authorisation route with the required safeguards.
For most small businesses, the risky area is recruitment, employee monitoring, credit scoring and AI-driven eligibility decisions. For example, using software to sort applicants into priority groups may be manageable if a trained manager reviews each applicant meaningfully. Automatically rejecting every applicant below a score without proper human review is much riskier. If you're considering systems like this, read our guide on automated decision-making and GDPR.
What Documents And Contracts Do You Need For Profiling?
Your documents should match what your business actually does. A generic privacy notice that says you may use data for marketing won't usually be enough if you're scoring people, enriching profiles, using AI recommendations or making automated decisions.
- Privacy Policy: Explain the profiling you do, the categories of data used, the purposes, lawful basis, sources of data and people's rights. For guidance, see our complete guide to Privacy Policies.
- Cookie Notice And Consent Mechanism: If you use analytics, pixels or advertising cookies, explain them clearly and get valid consent where required.
- Consent Forms: If you rely on consent, record what the person agreed to and make withdrawal easy. See our advice on GDPR consent forms.
- Records Of Processing Activities: Record the profiling activity, purpose, lawful basis, data categories, recipients, retention and safeguards.
- Legitimate Interests Assessment: If you rely on legitimate interests, document your purpose, necessity and balancing assessment.
- Data Protection Impact Assessment: If the profiling is high risk, complete a Data Protection Impact Assessment before you start.
- Data Processing Agreements: If a platform, agency or AI provider processes personal data for you, put proper data processing agreements in place.
- Is A DPIA Required? Do one for systematic and extensive profiling with significant effects, large-scale profiling, large-scale special category data or large-scale monitoring.
- Have We Reviewed Bias, Accuracy And Retention? Test outputs, remove inaccurate data, minimise fields and delete old profiles.
Here is a simple worked example. A hypothetical online training company has 8,000 customer records. It wants to score leads from 1 to 100 using course history, job title, email engagement and website visits. Anyone scoring over 70 will receive a sales call, and anyone under 20 will receive fewer emails.
The business is using personal data, and the scoring is automated. This is marketing, not a legal or similarly significant decision, so Articles 22A to 22D are unlikely to be triggered. However, UK GDPR and PECR still apply. The business decides to rely on legitimate interests for CRM scoring, completes a legitimate interests assessment, updates its Privacy Policy, adds clear unsubscribe and objection routes, checks cookie consent for website tracking, and limits lead scores to 12 months before recalculation or deletion. It also makes sure its CRM contract includes data processing terms and security obligations.
Now change the facts. The same score automatically blocks customers from accessing a professional certification course that they need for work, with no human review. That could be a significant solely automated decision. The business would need to assess Articles 22A to 22D, add safeguards, allow human intervention and likely complete a DPIA before launching.
Key Takeaways
- UK GDPR doesn't regulate profiling a company as such. It regulates profiling of living individuals, including customers, employees, sole traders, directors and business contacts.
- Common small business tools can involve profiling, including CRM segmentation, website analytics, lead scoring, recruitment software, employee monitoring and AI recommendations.
- Consent isn't always required, but you need a lawful basis. PECR consent may be needed for cookies, pixels and some electronic marketing.
- Public professional profiles and Companies House data can still contain personal data. Public availability doesn't remove your transparency and fairness obligations.
- Solely automated significant decisions are now handled under UK GDPR Articles 22A to 22D, with safeguards such as human intervention and a right to contest the decision.
- High-risk profiling needs stronger documentation, often including a DPIA, tailored privacy notices, processor contracts and checks for bias, accuracy and retention.
If you'd like help with reviewing your profiling activities, updating your privacy documents, checking your lawful basis or preparing data processing agreements, you can reach us at 08081347754 or team@sprintlaw.co.uk for a free, no-obligations chat.
Build privacy controls around the real data flow
What should the business document next?
Map the purpose, roles, lawful basis, notices, processor terms, retention, rights requests and incident response before relying on a policy alone.







