Squiid vs Merge
Merge gives a B2B product one normalised API across hundreds of customer systems. Squiid gives a project one key across the infrastructure it runs on. Both say unified API and mean opposite things.
Short answer. Merge is for products that must read and write data in whatever HR, CRM or accounting system each customer uses, behind one normalised data model. Squiid is for a project that needs a database, host, email sender, storage and models of its own.
The one-line difference. Merge unifies the systems your customers already run. Squiid unifies the vendors you buy from to run your own product.
Side by side
| Merge | Squiid | |
|---|---|---|
| What it covers | A unified API across customer systems: HR, recruiting, CRM, accounting, ticketing and file storage, with normalised models, syncs and webhooks. | The services your app runs on: models, databases, hosting, email, SMS, storage, auth, search, observability and payments, provisioned and billed. |
| Non-AI services | Entirely non-AI, but these are your customers business systems, not your infrastructure. | Yes. Postgres, object storage, email, phone numbers, error tracking. |
| Provisioning of accounts | It authorises against systems your customers already pay for. It does not provision anything for you. | Yes. Squiid opens the account and holds the credential. |
| Key custody and rotation | Yes, for customer connection credentials, managed per linked account. | Squiid holds and rotates the provider keys. Your project carries one gateway key, worthless outside it. |
| Billing model | Its own plans, commonly per linked account, as of their public docs. Customer systems are billed to customers. | Prepaid credits at par: $1 of credit is $1 at the provider. Subscriptions pass through at list price. |
| Fee | No fee on provider usage. You are buying normalisation and maintenance. | Free $0/mo + 15% at top-up. Solo $19.97/mo from 8%, falling to 3.5% with spend. Team from $29.97/seat, one point lower. Custom 3% or less. |
| One invoice PDF | A Merge invoice, plus every infrastructure bill you already had. | Yes. One monthly PDF: credits, plan, subscriptions. |
| Spend caps and pause-on-zero | Sync and linked-account limits. No spend balance across vendors. | Yes. Alerts at 75, 90 and 100%, capped auto top-up, pause at zero. |
| Works with coding agents | Yes, in that an agent can call a normalised API, though it is aimed at B2B integrations. | Yes. One variable and one base URL convention for every service. |
| Self-host option | No. Hosted platform, as of their public docs. | No. Hosted only. Leaving is a transfer where the provider supports one, a documented export otherwise. See handover. |
| Best for | B2B software that must integrate with whatever system each customer uses. | Projects that need infrastructure bought, credentialed and billed, rather than end-user SaaS accounts connected. |
When to pick Merge instead
Pick Merge when the integration surface is enterprise software you do not control. Normalising twenty HR systems into one data model is a miserable engineering project, and buying it is usually correct.
Pick it when the deal depends on it. In B2B sales integrations are often a contract condition, and a unified API turns a quarter of work into a configuration change.
Pick it when maintenance is the real cost. Those upstream APIs change, deprecate and break, and somebody has to keep up. Better them than you.
When Squiid is the better fit
Squiid is the better fit for the stack under your own product. Merge assumes you already have somewhere to put the data it syncs. Postgres, a host, object storage, an email sender and an auth tenant are five vendor relationships Squiid opens, holds and bills so you never do.
It is also the better fit for spend that scales with usage. A per-linked-account bill is predictable; model tokens, database compute, email volume and SMS are not, which is why they run against a balance that pauses at zero.
And it is the better fit for the number of credentials in your repository. One gateway key worthless outside the gateway beats a drawer of working provider secrets, however tidily stored.
Use both
A B2B product with a real integrations story runs something like Merge for customer systems and something like Squiid for its own infrastructure. One is inbound data from customers, the other outbound spend on vendors.
Questions people ask
Is Squiid a unified API like Merge?
A unified access layer, but over vendors you buy from rather than systems your customers run. One key and one base URL convention reach every service; there is no normalised data model, because a database and an SMS API have nothing to normalise.
Could Merge connect me to Supabase or Twilio?
Not the category it serves. Merge is built around business systems: HRIS, ATS, CRM, accounting. Infrastructure vendors are what the Squiid catalogue covers.
Do I need both?
Only if you sell to businesses and integrate with their systems. A consumer app or internal tool needs the infrastructure half and nothing else.
Which one is more likely to surprise me on cost?
The infrastructure half, almost always. Linked-account pricing is predictable; an agent in a retry loop against a model API is not. Which is why Squiid is prepaid and pauses at zero.
Does Squiid normalise anything across providers?
Only the access convention: one key, one base URL shape, one invoice. The provider APIs stay exactly as their own SDKs expect, which lets an agent use the documentation it already knows.
Two kinds of unified.
Only one sends the bill.
Squiid unifies the vendors your product buys from. One key, one monthly invoice PDF.