Addingwell is the main European alternative to Stape for hosted server-side Google Tag Manager. Its pricing model differs in ways that matter more than the headline number.
At a single site and the same volume it costs more than Stape. Run several sites and it costs considerably less. The add-on pricing and the overage model drive that, not the plan table.
All figures below were checked on addingwell.com/pricing in August 2026. Pricing changes, so verify before you commit.
Addingwell pricing plans in 2026
| Included requests/month | Monthly price | Overage per 100K |
|---|---|---|
| 100K (free plan) | €0 | - |
| 1M | €90 | €2.30 |
| 5M | €120 | €2.00 |
| 12M | €210 | €1.75 |
| 25M | €360 | €1.55 |
| 50M | €640 | €1.40 |
| 100M | €1,190 | €1.30 |
Above 100M requests you contact them for a high-volume plan.
Prices are in euros and there is no annual billing option listed, so every plan is month to month at list price. If you are comparing against a dollar-priced competitor, do the conversion yourself at today's rate rather than trusting a comparison table someone wrote last year.
The free plan never expires
Addingwell puts this plainly: instead of a time-limited trial, they offer a permanent free plan up to 100,000 requests per month, framed as a way to understand your own request consumption before committing.
The hardest thing about pricing a server-side setup is that nobody knows their request count until they measure it, and a request is not the same thing as a GA4 event. A free plan that runs indefinitely at 100K lets you measure your actual consumption on real traffic instead of estimating it.
The free plan includes a custom domain and automated tag health monitoring, so it is a working setup rather than a crippled demo.
What counts as a request
Addingwell's definition is close to Stape's, and it is the number your bill is built on:
"We only count incoming requests sent to your Google Tag Manager container. This includes downloads of scripts like gtm.js, analytics.js, gtag.js, but not outgoing requests."
Two things follow.
Script loads count. As with any request-based sGTM host, loading your GTM script through the container consumes quota before any conversion event fires. Budget from measured requests, not from a GA4 event count.
Fan-out does not multiply. Their own example: one request into the container, routed onward to Google Analytics, Meta CAPI, BigQuery, and TikTok, is billed as one request. Adding destinations does not raise your bill. That is the correct way to price it, and worth checking on any vendor you evaluate.
Overage is billed per 100K
Most hosted sGTM pricing is tiered: cross the ceiling and you move up a plan. Addingwell bills the excess instead, per 100,000 requests, at a rate that falls as your plan grows: €2.30 per 100K at the 1M tier down to €1.30 per 100K at 100M.
Their FAQ says the extra is calculated at the end of the billing cycle and added to the invoice. Unused volume does not roll over.
For traffic that spikes, this is meaningfully better than a hard tier. A Black Friday month that doubles your requests costs you the overage on the extra, not a permanent upgrade you then have to call support to unwind. If your traffic is seasonal, model both shapes before assuming the cheaper list price is the cheaper year.
Add-ons, and why they decide multi-site pricing
Three add-ons, all monthly:
| Add-on | Price |
|---|---|
| Container (additional website) | €30 |
| Domain (additional domain) | €10 |
| Region (multi-region infrastructure) | €20 |
The container add-on is the one that changes the arithmetic.
The common alternative model, used by Stape, is that one container equals one site and each site needs its own full subscription. On that model five sites at the 5M tier is five full plan prices. On Addingwell's model it is one plan plus four €30 add-ons.
That difference compounds fast for multi-brand retailers, agencies running client containers, and anyone with separate regional domains. If you are in that position, this add-on is probably the deciding factor in the whole comparison, well ahead of the base price.
Before you model it, check with Addingwell whether request volume is pooled across containers or allocated per container. The pricing page prices the add-on but does not settle that question, and the answer changes the maths considerably.
Addingwell is a Didomi company now
Didomi acquired Addingwell in 2025. Addingwell's own site describes the result as server-side tagging combined with a consent management platform, with consent travelling alongside the data.
Two practical consequences.
If you are already paying Didomi for consent management, or are about to buy it, the bundle is a genuine argument. Putting consent signals and tagging infrastructure under one vendor removes a real integration seam, and that seam is where most consent-mode implementations break.
If you have no interest in Didomi's CMP, you are buying a product whose roadmap now serves a consent-management strategy. That is not a criticism, but know what you are buying into.
When Addingwell is the right call
It is the right answer when you want hosted sGTM in Europe, you run more than one site, and your traffic is uneven enough that overage billing beats tier jumps. It is a stronger answer again if Didomi consent management is already in your stack or on your shortlist.
Like Stape, it is a server-side hosting and tagging product. It does not include an analytics interface, mobile measurement, or attribution reporting. Price the rest of the stack separately.
How it compares
Against Stape. Stape is cheaper for a single site: $83/month for 5M requests against Addingwell's €120. Stape also offers 20% off for annual billing, which Addingwell does not. Addingwell wins on multi-site (€30 per extra container versus a full subscription per site), on overage handling (billed, not forced upgrades), and on the consent bundle. Full breakdown: Stape pricing.
Against building on Google Cloud. By Stape's own published figures, a minimum production Cloud Run setup runs around $120/month for three servers before logging, which they estimate at roughly $100 per 500,000 requests. Managed hosting wins on cost for most teams. Build yourself when you need a custom architecture or specific residency guarantees.
Against Hardal. A different shape rather than a cheaper competitor. Hardal covers the same hosting layer, then includes web and mobile analytics, marketing attribution, ETL and data governance, and unlimited server-to-server destinations in a single product. Managed hosting runs in the US, UK, EU, EMEA and Asia, close to your users for latency or in a specific country for residency, and it can be self-hosted on your own infrastructure. Plans start at $249/month for 5M events.
That is more than €120, and it is not a like-for-like comparison in either direction. Addingwell's 5M is requests, including script loads; Hardal's 5M is events. And Hardal's price covers the analytics and attribution layer Addingwell expects you to buy elsewhere. If all you want is hosted sGTM in Europe with clean multi-site economics, Addingwell is the better and cheaper fit, and you should use it. If you are currently paying for hosting plus analytics plus a mobile measurement partner, compare the stack totals rather than the hosting line.
Full feature-by-feature breakdown: Hardal vs Addingwell. Current Hardal pricing: usehardal.com/pricing.
For another managed collection model, see the JENTIS pricing guide.
The short version
The add-ons and the overage model decide this one, not the plan table.
The free plan is permanent, so it is the cheapest honest way to find out your real request count. Overage is billed per 100K rather than forcing an upgrade, which suits spiky traffic. Extra containers cost €30 instead of a full subscription, and if you run several sites that is the whole decision. There is no annual discount. And it is a Didomi company now, either a significant plus or simply neutral depending on your consent stack.
Measure your requests on the free plan first. Then price it.
Running server-side tracking across several brands, or across web and mobile together? See how Hardal compares →