raptorservices.com

Working with data means turning raw records into understanding. In practice, that breaks into five repeatable activities: collecting data, cleaning it, exploring it, visualizing it, and interpreting what the results do and do not support. Data visualization and statistics are two halves of the same job — statistics tells you whether a pattern is real and how uncertain it is, while visualization shows you the shape of the pattern and communicates it to others. You do not need a math or programming background to start; you need a question, a small dataset, and a tool simple enough that you spend your time thinking about the data rather than the software.

The Five Core Activities of Data Work

Most data projects, from a personal budget spreadsheet to a public health dashboard, move through the same stages.

1. Collecting

You gather observations: survey responses, website logs, sensor readings, government tables, or a hand-built spreadsheet. The key decision here is what counts as one row (a person? a day? a transaction?) and what each column measures. Getting this "unit of observation" wrong causes problems that no amount of later analysis can fix.

2. Cleaning

Real data arrives messy. Cleaning means handling missing values, fixing inconsistent categories ("USA," "U.S.," "United States"), correcting types (a date stored as text), and removing duplicates. Beginners are often surprised that this is the most time-consuming step. It usually is.

3. Exploring

Before making charts for others, you look for yourself. What is the range of each variable? Are there outliers? How are two variables related? Simple summaries — counts, averages, minimums, maximums — and quick scatterplots answer most early questions.

4. Visualizing

You encode values as position, length, color, or size so that patterns become visible. A good chart answers one question clearly. A bad chart hides the answer behind decoration or distorts it through a misleading axis.

5. Interpreting

You decide what the pattern means, how confident you should be, and what alternative explanations exist. This is where statistics and careful reasoning matter most.

Visualization vs. Statistics: How They Complement Each Other

These are not competing approaches. They answer different questions about the same data.

A practical rule: visualize to discover, model to confirm, visualize again to communicate. A scatterplot might reveal that one region behaves completely differently from the rest; a statistical model then tests whether that difference holds up; a final chart shows the finding to an audience.

Beginner-Friendly Tools and Formats

You can start with tools you already have.

• Spreadsheets (Excel, Google Sheets): Best for datasets under a few thousand rows. Built-in chart types cover bar, line, scatter, and pie. Learn to sort, filter, and use pivot tables.
• Chart types to master first: bar charts for comparisons, line charts for change over time, scatterplots for relationships, and histograms for distributions. These four cover most everyday questions.
• Simple code options: If you want to go further, R (with ggplot2) and Python (with matplotlib or plotly) are common. Both have large free learning communities. Start with one, not both.
• Design principles that matter more than the tool: label your axes, start bar charts at zero, avoid 3D effects, use color to encode meaning rather than decoration, and put the most important comparison in the most prominent position.

A Realistic Starting Path

If you have no data background, this sequence works:

1. Pick a question you actually care about. "How has my city's rent changed over ten years?" beats a generic tutorial dataset.
2. Find a small, public dataset. Government open-data portals and statistical agencies publish free tables.
3. Load it into a spreadsheet and clean it. Fix types, remove duplicates, note missing values.
4. Make three charts. One bar, one line, one scatter. Write one sentence under each describing what you see.
5. Ask what could be misleading. Is the sample representative? Is the time range fair? Could a third factor explain the pattern?
6. Repeat with a slightly harder question. Add a second variable, or try a simple statistical summary like a correlation or a group comparison.

Expect the first project to take longer than you think, mostly in cleaning. That is normal, not a sign you are doing it wrong.

What Data Can and Cannot Answer

Data can describe what happened, compare groups, estimate relationships, and quantify uncertainty. It cannot, on its own, establish causation without a proper study design, tell you what you should value, or compensate for a biased sample. A dataset collected from volunteers will not represent the general population no matter how sophisticated the analysis. Treat every result as "what this data suggests under these conditions," not as a final verdict.

Where to Go Next

FlowingData (flowingdata.com) focuses on data visualization and statistics for people who want practical, well-designed charts rather than academic theory. It is a reasonable place to browse examples, see how real datasets are turned into clear graphics, and pick up habits you can apply in your own work. Pair it with one spreadsheet tutorial and one public dataset, and you have everything you need for a first project.

The short version: working with data is a craft of asking clear questions, cleaning messy inputs, looking before you model, and communicating honestly. Start small, start visual, and let the statistics grow as your questions get harder.

Customer data is the record of how individual customers behave and transact with your business — pages browsed, emails opened, products bought in-store or online. You use it by unifying those records into one platform, shaping them into audiences, and feeding those audiences into personalisation such as product recommendations, triggered emails, and merchandising. This applies to any business whose customers interact across more than one channel; if all your customer activity happens in a single system you rarely touch, the payoff is smaller.

What counts as customer data

Customer data is behavioural and transactional information tied to a person or account. It is not the same as aggregate business metrics.

The distinction matters because personalisation depends on knowing who did something, not just how many people did it.

Where customer data comes from

Customers generate data at every touchpoint. Common sources:

• Website browsing — pages viewed, categories explored, products compared
• Email engagement — opens, clicks, and which products drew attention
• In-store purchases — transactions rung up at physical locations
• Online transactions — orders, cart contents, purchase history

A retailer with both physical stores and an online shop accumulates these in separate systems by default. Imerco, for example, unified purchase data from 165+ physical stores and its online shop to build audiences from both.

Why unifying across touchpoints matters

Data sitting in separate systems gives you a partial view. A customer who browses online but buys in-store looks like two different people — or like a browser who never converts.

Bringing sources together in one platform lets you shape them into audiences that reflect who your customers really are. That accuracy is the input to everything downstream: if the audience is wrong, the recommendations and triggers built on it are wrong too.

XL-BYG previously had product directors manually choosing recommendations across 200,000+ products. After unifying data in a CDP with a personalisation engine, recommendations became automatic and real-time. Helene Birk's summary: "Our recommendations are now much more accurate without spending any time on them, and they have lifted our conversion rate considerably."

From unified data to personalised experiences

Once data is unified, it feeds personalisation in several concrete ways.

Recommendations

Personalise homepages, category pages, and product pages with suggestions grounded in each visitor's browsing and shopping behaviour. The same logic applies inside email: product recommendations for each subscriber, built into your existing email platform.

Triggers

Send automated emails the moment behaviour calls for it — an abandoned cart, a price drop — turning high-intent moments into conversions. Imerco combined behavioural email triggers with unified audiences and reported a 74% open rate on triggered emails and €1.25 revenue per email.

Merchandising

Choose specific products to boost, or build curated product feeds shown across websites and emails to the right audiences. This is where unified data and manual judgement work together rather than replacing each other.

Reported outcomes

Raptor's published results from customer deployments include:

• 102% increase in email revenue
• 52% purchase completion rate
• +273% conversion rate with recommendations
• +37% average order value
• +66% more unique products per order
• ~10x increase in click-to-open rate (Imerco)

Treat these as vendor-reported figures from specific customers, not guarantees. Your results depend on your data volume, channel mix, and how well your audiences are built.

What to check before investing

• Do you have data in more than one place? If website, email, and store data live in separate systems, unification is the prerequisite for everything else.
• Do you have enough behavioural volume? Recommendations and triggers need activity to learn from.
• Which use case comes first? Abandoned-cart triggers are typically simpler to launch than full on-site recommendation engines.
• How will you measure it? Conversion rate, average order value, and email revenue are the metrics referenced in the results above.

Pricing and plan details are not covered here — check the vendor's pricing page for current terms.

A Customer Data Platform (CDP) is a system that brings customer data from every touchpoint — website visits, email engagement, in-store purchases — into one place, then shapes that data into audiences you can act on. You get value from it when you connect all relevant sources and use those audiences to personalise websites, emails, and product recommendations. Raptor describes its product as an "Agentic Customer Data Platform" built around exactly this: unifying customer data, building precise audiences, and automating personalised recommendations across emails and websites.

How a CDP turns scattered data into usable audiences

Customers generate data at every interaction. A CDP collects that data and organises it so it reflects who your customers actually are, rather than fragments spread across tools.

The mechanism has three parts:

1. Collection — data flows in from browsing, email activity, and in-store shopping.
2. Unification — records are brought together into one platform so a single customer isn't split across systems.
3. Audience building — the unified data is shaped into groups that reflect real behaviour, which you can then target.

The output is not a report. It's a set of audiences and signals that other systems can act on immediately.

What you can actually do with it

Once data is unified, a CDP supports activation across channels. Raptor's platform lists four concrete capabilities:

The common thread is activation: the data doesn't sit in a dashboard, it drives what each customer sees.

What results look like in practice

Raptor cites two customer examples that show how unified data and audiences translate into outcomes.

XL-BYG previously had product directors manually choosing recommendations across 200,000+ products. After adopting Raptor's personalisation engine and CDP, recommendations became automatic and real-time, and customer audiences sharpened targeting across the range. Helene Birk said recommendations are "much more accurate without spending any time on them" and have lifted conversion rate considerably.

Imerco unified purchase data from 165+ physical stores and its online shop, then combined those audiences with behavioural email triggers. Email became a core sales channel delivering personalised messages rather than a generic newsletter.

Reported results across Raptor's customers include a 102% increase in email revenue, a 52% purchase completion rate, +273% conversion rate with recommendations, and +37% average order value.

How a CDP differs from a generic data tool

A generic analytics or data tool helps you understand what happened. A CDP is built around activation — it focuses on pushing unified data into the channels where customers actually engage.

That distinction matters when you evaluate options. Ask whether the platform:

• Connects to your website and email systems directly, or only exports data
• Builds audiences you can target, or only segments you can view
• Automates recommendations and triggers, or leaves execution to you

Raptor's positioning is explicitly on the activation side: recommendations and triggers built into existing email platforms and websites.

What determines whether it works for you

Results depend on two conditions:

• Data quality — unified data is only as good as what flows in. Incomplete or inconsistent sources weaken audience accuracy.
• Source coverage — connecting all relevant touchpoints (web, email, in-store) is what makes audiences reflect real behaviour. Imerco's results came from unifying 165+ stores with its online shop, not from one channel alone.

If your customer data is fragmented across channels and you want to act on it rather than just report on it, a CDP is the category to evaluate. If you only need reporting, a standard analytics tool may be enough.

Personalisation in a customer data platform (CDP) means using each customer's own data and behaviour to decide what they see — which products appear on a homepage, which recommendations land in an email, and which message gets triggered by a specific action. It works when three things are in place: customer data from every touchpoint is unified in one place, that data is shaped into audiences, and those audiences drive automated recommendations and messages across websites and email. If your data still lives in separate systems per channel, personalisation stays shallow — you can personalise within a channel, but not across the customer's actual journey.

Why unified customer data is the starting point

Customers generate data at every touchpoint — browsing your website, opening an email, shopping in-store. A CDP's job is to bring that together in one platform and shape it into audiences that reflect who customers really are.

That unification is what separates real personalisation from surface-level tactics. If purchase data from physical stores never meets online browsing data, you can't tell that a loyal in-store buyer is browsing a category online without converting. Once the data is unified, the same audience can follow the customer from a website recommendation into an email.

Imerco's case shows the pattern: purchase data from 165+ physical stores and the online shop is unified in a CDP, and precise audiences are built from both. Those audiences then feed behavioural email triggers, so email delivers a personalised message to each customer rather than a generic newsletter.

The four common personalisation outputs

These outputs are not alternatives — they usually run together. A visitor sees personalised recommendations on-site, receives a triggered email if they abandon, and the merchandising team can still push curated feeds to specific audiences.

How behavioural triggers turn intent into conversion

Triggers are where personalisation becomes time-sensitive. Instead of a scheduled send, the email fires the moment customer behaviour calls for it — an abandoned cart, a price drop, or another high-intent signal. The mechanism is straightforward:

1. Input: a behavioural event (cart abandoned, price dropped on a viewed product).
2. Action: the CDP matches the event to the customer and the relevant audience, then the email platform sends the personalised message.
3. Expected result: the customer receives a message tied to what they were just doing, at the moment they were doing it.

Imerco's triggered emails reached a 74% open rate, with €1.25 revenue per email and roughly a 10x increase in click-to-open rate. Those numbers reflect triggered, personalised messages rather than batch newsletters.

What results personalisation is associated with

Raptor's published results give a sense of the outcomes teams report:

• 102% increase in email revenue
• 52% purchase completion rate
• +273% conversion rate with recommendations
• +37% average order value
• +66% more unique products per order
• -64% lead price in META

XL-BYG's story explains the operational side: the retailer previously relied on product directors to manually choose recommendations across 200,000+ products. With a personalisation engine and CDP, recommendations are now delivered automatically and in real time, and customer audiences sharpen targeting across the range. As Helene Birk put it, recommendations are "much more accurate without spending any time on them," and they lifted conversion rate considerably.

When this approach fits — and when it doesn't

Personalisation through a CDP makes sense when you have customer data spread across channels (web, email, in-store) and enough product or content volume that manual curation doesn't scale — the XL-BYG case, with 200,000+ products, is the clear example. It also fits when you want triggered messages tied to real-time behaviour rather than scheduled campaigns.

It fits less well if your data is already unified and your catalogue is small enough that manual merchandising works, or if you only need personalisation inside a single channel and have no ambition to connect touchpoints. The value comes from the connection between channels, not from any single output.

If you want to see how this works on your own data, Raptor offers a personalised demo, and pricing details are available on their pricing page.

Recommendations in a customer data platform (CDP) are product or content suggestions generated automatically from each customer's browsing and shopping behaviour, then delivered on websites and in emails. They work when your customer data from every touchpoint — website, email, and in-store — is unified in one place, so the platform can match the right product to the right person at the moment they are most likely to act. If your data sits in separate tools and no single system knows what a customer viewed, bought, or abandoned, recommendations will stay generic.

What "Recommendations" Means Inside a CDP

A recommendation is not a manually curated list of bestsellers. In a CDP it is an output of the platform's understanding of each individual:

• Input: behavioural signals (pages viewed, products browsed, past purchases, email engagement) plus audience membership built from unified customer data.
• Action: the platform scores and selects products for that specific visitor or subscriber.
• Output: personalised placements on homepages, category pages, and product pages, or product blocks inside emails.

The distinction matters because it changes who does the work. Instead of a merchandiser choosing what to show across a catalogue, the platform generates suggestions per person and updates them as behaviour changes.

How Customer Data Feeds Recommendations

Recommendations are only as good as the data behind them. The platform's role is to unify data that customers generate at every touchpoint and shape it into audiences that reflect who those customers really are.

A practical example: a retailer with both physical stores and an online shop. Purchase data lives in the store systems, browsing data lives on the website, and email engagement lives in the email tool. Once these are unified, a customer who bought in-store but browses online can receive recommendations informed by both — not just by whatever the website happens to know.

This is the mechanism behind the XL-BYG case: the retailer previously relied on product directors to manually choose recommendations across 200,000+ products. With a personalisation engine and CDP, recommendations are now delivered automatically and in real time, and customer audiences sharpen targeting across the range. As Helene Birk put it, "Our recommendations are now much more accurate without spending any time on them, and they have lifted our conversion rate considerably."

Website Recommendations vs. Email Recommendations

Both use the same underlying customer data, but they serve different moments.

Website recommendations suit visitors who are already engaged and deciding what to look at next. Email recommendations suit subscribers who are not on your site at that moment, so the email itself has to carry the personalisation.

Combining Recommendations With Behavioural Triggers

Recommendations become more powerful when they are attached to a moment of intent rather than a fixed schedule. Behavioural triggers send automated emails the moment customer behaviour calls for it — for example, an abandoned cart or a price drop — and recommendations fill those emails with the specific products that matter to that person.

The logic: a trigger identifies when to send, and recommendations decide what to show. An abandoned-cart email without recommendations is a reminder; with recommendations it can surface the abandoned item plus relevant alternatives.

How to Tell Whether Recommendations Are Working

Track metrics that connect recommendations to revenue, not just engagement:

• Conversion rate — the platform reports +273% conversion rate with recommendations.
• Average order value — reported at +37%.
• Unique products per order — reported at +66% more unique products per order, a sign recommendations are broadening baskets rather than repeating the same items.
• Email revenue — reported at 102% increase in email revenue.
• Revenue per email — Imerco reports €1.25 revenue per email and a ~10x increase in click-to-open rate.
• Triggered email performance — Imerco reports a 74% open rate on triggered emails.

The Imerco case shows the combined effect: purchase data from 165+ physical stores and the online shop is unified into precise audiences, then paired with behavioural email triggers so each customer receives a personalised message rather than a generic newsletter.

What to Check Before Choosing a Platform

• Does it unify data across all your touchpoints, including offline? Recommendations built on partial data will underperform.
• Can it serve both website and email placements from the same customer profiles, or will you need to stitch tools together?
• Does it support behavioural triggers so recommendations can be tied to intent moments like cart abandonment or price drops?
• Can merchandisers still intervene — boosting specific products or building curated feeds for the right audiences — when business priorities require it?
• What results does it report, and on which metrics? Conversion rate, AOV, and revenue per email are the ones that connect to commercial outcomes.

If your current setup requires manual product selection across a large catalogue, or your website and email personalisation run on separate data, that is the clearest signal a CDP-driven recommendation engine would change your results.