For in-house teams and founders
Know which ads make you money, not just revenue.
If your store runs on Shopify or WooCommerce, 1ads joins your ad accounts with your store, so every campaign is judged on profit after COGS, refunds and refused cash-on-delivery parcels. It checks spend every hour, 12 hours or day, depending on your plan and add-ons, and proposes changes you approve.
The numbers your ad platforms do not show you
- Profit on ad spend (POAS) per campaign and channel, after COGS, shipping, payment fees, refunds and refused COD parcels, once your product costs are mapped (we need COGS on at least 80% of sales).
- New versus returning customers: what you pay to acquire a new customer, and when they pay you back (30, 60, 90 and 180 days).
- The cash-on-delivery gap. For a Romanian shop spending €10k a month with 30% cash on delivery, our model expects about 22 refused parcels a month, about $176 a month of shipping and handling, and platform revenue about 2% higher than what you actually collect. This corrects what you measure; the direct saving from fewer refusals is smaller, and we count it only once your pilot measures it. If your store runs on Shopify or WooCommerce, 1ads shows you your own number in week two. (Assumption: 7% of COD orders refused; PayU Romania reports about 11% of COD orders returned or refused; there is no national statistic.)
- A confidence grade on every number, from A (a recent controlled test) to D (estimate only), and the cheapest next step to improve it.
What 1ads is likely to return, by size of business
We will publish modelled results for typical businesses only after the first pilots have measured how many AI credits real teams use, because credits are part of the cost. Until then, the ROI calculator lets you enter your own usage and shows the conservative case first.
What our model already says, on current assumptions: start small and start with a pilot. For a single-marketer shop, Starter with one Guardian is the configuration to test; Growth adds cost a one-person team rarely recovers. Western-EU teams are carried mostly by the hours they save, so lower wages (Central-Eastern Europe, Romania) make the case harder. A shop spending about €2k a month on one channel does not pay back. Large brands may find that procurement does not pass yet, because we have no SOC 2.
Not sure it pays off? Start with a 90-day pilot at a fixed price, and leave without penalty if the measured result falls below your break-even. Request a pilot
Hourly checks across all your networks, judged on profit, with changes you approve
- Every hour: spend, pacing, delivery, tracking breaks, invalid clicks. No strategic budget moves on hourly data.
- Every 12 hours: re-check with conversions corrected for reporting delay; "market or campaign?" label.
- Every day: the profit ledger, incrementality factors and recommendations.
- Prices: an hourly Guardian is $129/month, a 12-hour one $39 and a daily one $19 (549 / 179 / 89 lei on the Romanian price list). Pro includes 2 hourly and 6 twelve-hour automations (8 twelve-hour on the Romanian price list); an included 12-hour slot can run hourly for the price difference ($90 / 370 lei a month).
- What is not different: native rules already check often (Meta's automated rules run continuously, which third-party documentation puts at about every 30–60 minutes; Google Ads rules can run hourly; TikTok has automated rules), and rule tools such as Birch check as often as every 15 minutes. They act on their own and judge on platform metrics. Guardian checks every network you connect, judges on your targets and profit, and waits for your approval.
- Honest limits: LinkedIn and Reddit only report daily, so hourly checks are not offered for them. Meta allows at most 4 budget changes per ad set per hour.
Is your brand search really adding sales?
Research on eBay found brand-keyword ads produced no measurable short-term benefit, while another study shows competitors take 18–42% of clicks when a brand stops bidding on its name. Both can be true for different businesses, so we do not assume: 1ads helps you run a holdout (for example, pausing brand search in a set of Romanian counties for three weeks), estimates the cost of the test in advance, and records the result in your incrementality ledger. The test can cost you some sales; our calculator includes that cost. (Blake, Nosko & Tadelis, Econometrica 2015; Simonov, Nosko & Rao, Marketing Science 2018.)
Small accounts often cannot reach statistical certainty; we will say so and grade the result C or D rather than overclaim. Google's own conversion-lift studies need about $5,000 of spend.
Questions brands ask
Does 1ads work with Shopify and WooCommerce?
Yes, for orders, refunds and COGS. Shopify only stores the current cost of a product, so we take a daily snapshot from day one; WooCommerce has little native COGS, so we help you import costs during onboarding. Gomag, MerchantPro and PrestaShop are not connected yet.
Where do refused COD parcels come from?
From your store's order statuses, and later from courier data (FAN Courier, Sameday, Cargus, DPD) where you connect it.
Can I keep my agency?
Yes. Give them access to your workspace; they propose, you approve.
Why pay 1ads instead of using Meta's and Google's free automation?
Keep using them. Advantage+, Performance Max and AI Max are free, they use signals no outside tool can see, and 1ads does not bid instead of them. What they do not do is judge themselves: each platform optimises and reports on its own attribution, inside its own walls. 1ads sits above them. It joins ad data with your store and analytics, judges every channel on your targets and, where your store data allows, on profit after COGS, refunds and refused COD parcels, flags when tracking breaks or delivery stops, proposes moving budget between channels, and helps you test whether a channel really adds sales (for example with a brand-search holdout). If you advertise on one channel, your tracking is clean and you are happy with the results, you probably do not need us. Our own model says a shop spending about €2k a month on Meta alone does not pay back.
Does the AI change my campaigns on its own?
No. The AI can read, analyse, draft and propose. It has no permission to apply anything. A change runs only after an authorised person approves it. In the app, an approval needs multi-factor authentication within the last 15 minutes. From email or Slack, the approver confirms with a one-time code; approval links expire and work only once. During the closed access program there is no auto-approval at all, with no exceptions in any contract. Before an approved change runs, 1ads re-reads the live campaign; if something changed since the proposal, it goes back for approval. Hard limits, such as how much a budget may change in a day, apply even to approved changes. The trade-off is speed: at 3 a.m. a person has to approve, so tools with autonomous rules react faster.
How do you calculate ROI?
Two different things carry that name.
Your ads' return inside 1ads: we start from profit, not revenue. Profit on ad spend (POAS) is net revenue minus COGS, shipping, payment fees, refunds and refused COD parcels, divided by ad spend. We never add up conversions reported by different platforms. Each channel then gets an incrementality factor from tests and cited research, and every result shows a range and a confidence grade from A (a recent controlled test) to D (estimates only).
The value of paying for 1ads: the calculator adds what you can verify first (tools you cancel, then net hours saved × your loaded hourly cost), then problems caught earlier and a cautious range for ad efficiency, and subtracts the plan, add-ons, AI credits, the hours you spend setting it up and the cost of any holdout. It shows the net value and payback first, then value-to-cost, for the conservative and expected cases, with and without ad improvement.
Why closed access?
Three reasons. First, platform approvals: Google, Meta, LinkedIn and TikTok review apps like ours before granting full access, so channels arrive progressively, and we would rather onboard you when your channels work. Second, we want to measure results with every customer (hours, tools cancelled, one guided test) instead of guessing them. Third, contracts and invoices are manual during the closed access program: an Order Form you sign, invoices paid by bank transfer. So a person reviews each request, usually with a short call, and we may say "not yet" if we are not a good fit for you.
Where is my data stored?
Your account and customer data are stored in the EU, on our own on-premises Kubernetes cluster in a data centre in Romania (not a public cloud). We run the cluster ourselves, under our FirstIT brand. To answer a chat request or run an analysis, the excerpts needed for that task are sent to our AI provider, Anthropic, which processes them in the United States under EU-approved transfer safeguards (the mechanism is stated for each provider in the Sub-processors list). Images are generated by OpenAI, also in the United States, and transactional email is sent through Amazon SES (Amazon Web Services EMEA SARL, EU region). Under their commercial terms, neither AI provider trains its models on this data. Until zero data retention is confirmed, Anthropic deletes API inputs and outputs within 30 days, except content flagged by its safety systems (kept up to 2 years) and where the law requires; OpenAI may keep abuse-monitoring logs for up to 30 days. Tokens for the platforms we connect directly are encrypted in our vault and never sent to an AI model; channels connected through data partners are authorised with that partner. Each customer's data is kept separate. The sub-processor list, with locations, is public, and we give 30 days' notice before adding one. An EU-only AI processing option is being evaluated; it is not available yet.
Test it on your own numbers first
The calculator shows the conservative case first, with and without ad improvement. If it is below 1x, ask for a pilot.