Sandveld Home
Sandveld Home
How we use your data
What this dashboard reads from your accounts, what it keeps, and who can see it. If anything here doesn’t match what you expected, tell your account manager.
Which figure to go by
Different tools count the same thing differently. For each part of the business one source is the record – the source of truth – and figures from the others are labelled as theirs wherever they appear.
| What it measures | Source of truth | Also shown, labelled as theirs |
|---|---|---|
| Sales, orders and return on spend | ShopifyEvery sale, order and refund is recorded in the store first. Sales credited to ads or email are the store’s orders whose visit came from them, on last click. |
|
| Visits to the site | ShopifyThe store counts every visit, cookies or not. |
|
| Visits from Google search | Google Search ConsoleGoogle’s own record of every click from its results to the site. |
|
| Ad spend, impressions and clicks | Each ad platform – Google Ads, Meta, TikTok | Nothing else counts this, so there is nothing to weigh it against. |
| Emails sent, opened and clicked, and subscribers | The email platform – Klaviyo or Omnisend | Nothing else counts this, so there is nothing to weigh it against. |
| Where visitors get stuck on the site | Microsoft Clarity | Nothing else counts this, so there is nothing to weigh it against. |
| Followers, views and engagement on social | Each social platform | Nothing else counts this, so there is nothing to weigh it against. |
| Rankings, competitors and linking sites | DataForSEO | Nothing else counts this, so there is nothing to weigh it against. |
What we connect to
12 sources, each read once a night.
To report revenue, orders and which campaigns the store itself credits for them.
- Each order’s date, financial status, total and refunded amount
- The traffic source, UTM tags and landing page of the visit that led to the order
- Whether the order was that customer’s first – Shopify’s count of their orders, not who they are
- Each order’s order ID, to look up the channel Google Analytics recorded for it
- The products on each order: product and variant, quantity, price, discounts and refunds
- Each product’s title, type, collections and main image
- Each variant’s SKU, price, compare-at price and unit cost, and its stock at each location
- Each order’s number, totals, tax, discount codes, payments and their fees, refunds and the date of each
- The customer number on each order, to count repeat customers – never their name, email, phone or address
- Shopify’s own report of orders and total sales by shipping country and province – not any address
- Each order’s fulfilments (when sent, how many items) and returns (items, quantities and the return reason chosen – never the customer’s own note)
- The store’s open fulfilment orders: which items are still to send, from where, and whether any are on hold
- Daily totals for revenue, orders and refunds
- Daily revenue and order counts per campaign tag
- Daily orders and revenue from new and from returning customers
- For each campaign tag, its most common UTM source, medium and landing page
- Daily units, sales and refunds per product and variant, and per product by channel and campaign
- Product titles, types, collections and image links
- Stock per variant per location once a day, with its price, compare-at price and unit cost
- Each order and its lines: products, quantities, prices, discounts, tax, payments, refunds and the visit that led to it, for two years
- Each order’s customer number, where the store allows it
- Daily orders and total sales per shipping country, and per country and province
- Each order’s fulfilments and returns, for two years, and last night’s list of items still to send
Personal data: Each order’s Shopify customer number is kept, so repeat customers can be counted and each one opened in the store’s own Shopify admin. No customer names, email addresses, phone numbers or postal addresses are requested or stored.
To report site traffic, engagement and conversions, and where visitors came from.
- Daily totals for sessions, users, new users, engaged sessions, conversions and revenue
- The same totals broken down by acquisition channel
- Sessions by source and medium, to check ad tagging
- Page views and time engaged for each day’s busiest 100 pages, and sessions by device
- Product page views and add-to-carts per product, with the product’s name
- For each purchase, its order ID, channel and Google Ads campaign, to tell ad orders from organic ones in store revenue
- Those daily totals, exactly as read
- Each purchase’s order ID with its channel and Google Ads campaign – no visitor or customer details
Personal data: None. Google Analytics returns already-aggregated figures; no individual visitor is identified or retrievable.
To report what paid search cost and what the platform credits it with returning.
- Daily cost, impressions, clicks, conversions and conversion value per campaign
- Campaign names
- Those daily figures per campaign
Personal data: None. Google reports campaign-level totals; no individual person is involved.
To report what paid social cost and what the platform credits it with returning.
- Daily spend, impressions, clicks, purchases and purchase value per campaign
- The same figures split by Facebook and Instagram
- Campaign names
- Each ad’s destination link and URL parameters, to see which UTM tags it sends people with
- Those daily figures per campaign, and per campaign on Facebook and on Instagram
- Per campaign: the utm_campaign its ads use, and how many of its ads are tagged
Personal data: None. Meta reports campaign-level totals; no individual person is involved.
To report what email and SMS contributed.
- Daily counts of emails sent, and unique emails opened and clicked
- Daily order revenue Omnisend attributes to its messages
- Each newsletter’s name, subject, send time, figures and the segments it went to, and each automation’s name and daily figures
- Each newsletter’s opens by the hour in its first week, read once
- The names of the brand’s segments
- Daily counts of people subscribing and unsubscribing
- Once a month, the list of subscribed contacts – only to count it
- The brand’s name, to label the connection
- Those daily counts and totals, the newsletters’ and automations’ names and figures, and the subscriber count
Personal data: Counting subscribers reads the contact list once a month; it is counted in memory and no contact, email address or phone number is stored. Everything else is aggregated figures only.
To report what TikTok advertising cost and what the platform credits it with returning.
- Daily spend, impressions, clicks, completed payments and payment value per campaign
- Campaign names, objectives and statuses
- The ad account’s name, status and currency, to label the connection
- Those daily figures per campaign, and campaign names and objectives
Personal data: None. TikTok reports campaign-level totals; no individual person is involved.
To report how the site performs in Google Search, and to check its sitemap, indexing and structured data during SEO audits.
- Daily clicks, impressions and average position in Google Search, for the whole site
- The same figures for the top 250 search terms and top 250 pages
- The sitemaps submitted, and whether Google could read them
- For each key page in an SEO audit: whether Google has indexed it, and any problems with its structured data
- The daily search figures, exactly as read
- Sitemap and page findings, as part of the SEO audit they were read for
Personal data: None. Search Console reports only aggregated figures; the search terms carry no information about who searched them, and Google withholds rare terms that could identify someone.
To track whether AI answer engines mention and cite the brand when asked the questions its customers ask, and to crawl the site for SEO audits.
- Each week: the answers ChatGPT, Google (AI Overviews and ordinary results), Perplexity, Claude and Gemini give to the tracked questions, with the sources they cite
- During an SEO audit: the site’s public pages, as a search engine crawler would see them
- Each answer’s text and sources, whether it names the brand or a listed competitor, and whether it cites the site
- Weekly counts of answers, mentions and citations
- The crawl’s findings: broken links, duplicate titles and descriptions, pages missing image alt text
Personal data: None. The questions are written by Maverick about the brand and its market; no customer or visitor data is sent to DataForSEO or the AI engines.
To report how the brand’s own TikTok posts perform, beside its sales.
- Daily views, likes, comments, shares, profile views and follower changes for the account
- The account’s follower total
- Each post’s caption, link and lifetime figures
- The audience’s countries, ages and genders, as percentages
- Those daily figures and post figures, and the audience percentages
Personal data: None. TikTok reports totals and percentages; no individual viewer is identified.
To report how the brand’s own Instagram posts perform, beside its sales.
- Daily views, likes, comments, shares, saves, profile views and follows for the account
- The account’s follower total
- Each post’s caption, link and lifetime figures
- Followers’ countries, ages and genders, as totals
- Those daily figures and post figures, and the audience as percentages
Personal data: None. Instagram reports totals; no individual follower is identified.
To report how the brand’s own Facebook posts perform, beside its sales.
- Daily views, post engagements, Page views and follows for the Page
- The Page’s follower total
- Each post’s text, link, reactions, comments and shares
- Those daily figures and post figures
Personal data: None. Facebook reports totals; no individual is identified.
To show where visitors get stuck on the site, so it can be fixed.
- Each day’s totals: visits, and the share with rage clicks, dead clicks, quick backs, script errors, error clicks and excessive scrolling
- The same by page and by device type
- Average scroll depth and time engaged, where Clarity gives them
- Those daily totals and their breakdowns by page and device – no recordings and nothing about individual visitors
Personal data: None. The export gives counts only; recordings and heatmaps stay in Clarity.
Who can see it
- The people at Sandveld Home we have given a login. They see only Sandveld Home.
- Maverick staff working on your account, who can see the clients Maverick reports on.
- Nobody else. Your figures are never shared with other clients, sold, or used to build anything outside your own reports.
Where it is kept, and for how long
- In a private database run on Railway, reachable only by this dashboard. Traffic to and from it is encrypted.
- The keys that let us read your accounts are encrypted before they are stored, so a copy of the database alone would not give anyone access to your platforms.
- We keep the daily totals for as long as you are a client, because a report is only useful against last year. Ask us to delete them and we will.
- Where your store is connected, we also keep its orders one by one – going back up to two years before it was connected – so the store reports can be worked out from them: what each order held, its payments, discounts, refunds, fulfilments and returns. These are kept, and can be deleted, the same way.
- Access can be withdrawn at your end at any time – remove the app or revoke the key in the platform concerned and the nightly read stops immediately.
Claude
Your account manager can also ask Claude about your figures – to check a report or find what moved. Claude reads the same totals, tables and store reports you can see here.
No personal data is sent to Anthropic: customer numbers stay in the dashboard. Where a store report lists customers, Claude sees each one only as its place in the list.