The truth behind your numbers.
Every ad platform marks its own homework. The Honest Data Engine checks it against your store, so you can see which campaigns found new customers and which just took credit for loyal ones.
Shopify · Meta · Google · Runs behind every client account
Add up what the platforms claim. Then check the till.
For one DTC brand over 90 days, Meta and Google together claimed credit for 92% of everything the store sold, including customers who'd have come back anyway.
Every campaign, judged on new customers.
Spend, what the platform claims, what the store actually took, and what each new customer really cost.
| Campaign | Spend | Platform ROAS | Store ROAS | New customers | Cost per new | AOV |
|---|---|---|---|---|---|---|
| Meta · Advantage+ Shopping | £18,400 | 3.9× | 2.25× | 460 | £40.00 | £68.00 |
| Meta · Broad prospecting | £9,600 | 3.1× | 2.27× | 300 | £32.00 | £64.00 |
| Meta · Retargeting | £3,200 | 8.4× | 5.77× | 40 | £80.00 | £71.00 |
| Google · Shopping (non-brand) | £7,800 | 5.6× | 5.09× | 520 | £15.00 | £62.00 |
| Google · Performance Max | £6,400 | 6.2× | 5.36× | 400 | £16.00 | £66.00 |
| Google · Search (non-brand) | £2,900 | 4.4× | 3.52× | 145 | £20.00 | £60.00 |
| Google · Search (brand) | £1,100 | 22.1× | 19.73× | 22 | £50.00 | £70.00 |
| Blended | £49,400 | – | 3.80× | 1,887 | £26.18 | £65.82 |
What a new customer is really worth.
Cost per new customer only tells half the story. The engine follows every new customer for 12 months, so you know what you can afford to pay to win the next one.
| First order | Month 1 | 3 months | 6 months | 12 months |
|---|---|---|---|---|
| Jul 2025 | £64 | £81 | £96 | £120 |
| Aug 2025 | £66 | £83 | £98 | £123 |
| Sep 2025 | £67 | £82 | £97 | £121 |
| Oct 2025 | £65 | £80 | £95 | not yet |
| Nov 2025 | £71 | £86 | £99 | not yet |
| Dec 2025 | £69 | £84 | £97 | not yet |
Six questions, answered from your own data.
Spend from the platforms, sales from your store, joined every night.
Platform claims vs store truth
What Meta and Google say they drove, next to what your Shopify store actually took, campaign by campaign.
Real new customers
Every order checked against the customer's full history, so a lapsed customer coming back is never counted as new.
Cost per new customer
By channel, campaign and ad set, on first-touch and last-touch, so you see who finds new people and who picks up existing ones.
Incremental ROAS, estimated
Platform ROAS adjusted by tactic using published incrementality research, ready to be replaced by your own lift tests.
Customer lifetime value
What each month's new customers go on to spend over 1, 3, 6 and 12 months, so you know what a customer is really worth.
Launches and anomalies
Product launches and unusual days flagged automatically, so a big drop doesn't flatter or punish the wrong campaign.
| First product bought | Role | What we found |
|---|---|---|
| Product A | Acquisition | Wins the most new customers of any product |
| Product B | High value | £104 average spend in the first 90 days, 22% buy again |
| Product C | Loyalty | 58% of its revenue comes from returning customers |
Which products win new customers?
For one DTC pet brand, the engine showed that two products were doing the hard work of finding new customers, while a third was really a loyalty product. That changed which products lead the ads and which go to existing customers.
The same view works for any Shopify brand: find your gateway products, and stop spending prospecting budget on the ones people only buy once they already love you.
Where should the next pound go?
Media mix modelling looks at months of spend and sales together to estimate what every channel contributes, including the effects clicks can't see, like a YouTube campaign lifting search and direct sales.
Geo holdout tests prove it. We switch a channel off in a set of matched regions, keep it running everywhere else, and measure the difference in sales. It's the most reliable way to know what a channel truly adds, and the results feed straight back into the model.
- How much each channel really adds, beyond what it can track
- Response curves that show where extra spend stops paying back
- A recommended budget split for the next month or season
- Geo holdout and lift tests that measure true incremental sales
Connected once. Checked every night.
Connect
Read-only access to Shopify, Meta and Google Ads. Nothing is changed, and no campaigns are touched.
Reconcile
Orders are cleaned of refunds, cancellations, test and wholesale orders, and every customer is checked against their full order history.
Attribute
Sales are matched to the campaign, ad set and ad behind them, on first-touch and last-touch, next to what each platform claims.
Calibrate
Platform numbers are adjusted towards true incremental impact, then tightened over time with your own lift tests.
No names. No emails.
The engine never asks Shopify for a customer's name, email, phone number, address or IP. It works on order IDs, products, dates and values only.
It runs behind every client account, and the results feed straight into your reporting dashboard, so you see the engine's answers without learning a new tool.
Not another reporting tool. A second opinion on every number.
How the Honest Data Engine compares with the tools most brands already rely on.
| Honest Data Engine | Ad platform reporting | Web analytics | Self-serve attribution software | |
|---|---|---|---|---|
| Starts from your store's actual orders | ✓ | ✗ | Partial | ✓ |
| New customers proven against full order history | ✓ | ✗ | ✗ | Partial |
| Platform claims shown next to store revenue | ✓ | ✗ | Partial | ✓ |
| Cost per new customer by campaign | ✓ | ✗ | ✗ | ✓ |
| 12-month customer value by cohort and channel | ✓ | ✗ | ✗ | Partial |
| Incremental ROAS estimate by tactic | ✓ | ✗ | ✗ | Partial |
| Set up, checked and explained for you | ✓ | ✗ | ✗ | ✗ |
| Extra software subscription | Included | None | None | Monthly fee |
Compared by type of tool, based on typical setups. Individual products and configurations vary. Ad platform reporting means tools such as Meta Ads Manager and Google Ads; web analytics means tools such as Google Analytics 4.
The Honest Data Engine, answered.
What is the Honest Data Engine?
It's Honest Social's own measurement system. It joins Shopify orders with Meta and Google Ads spend to show what paid media actually added to your store: real new customers, cost per new customer, and how much of what the platforms claim was genuinely incremental.
Is it a separate product?
No. It runs behind every Honest Social client account and feeds your reporting dashboard. You get its answers as part of working with us.
Which platforms does it connect to?
Shopify, Meta Ads and Google Ads today, all with read-only access.
How is this different from what Meta and Google report?
Platforms count any sale that followed an ad, including customers who'd have bought anyway. The engine starts from your store's actual orders, checks whether each customer is genuinely new, and compares that with what each platform claims.
What did your ads actually add?
Book a call and we'll show you the engine on your own Shopify and ad data.