FAQ · 68 answers · 10 topics
Frequently asked questions about Vexio
Vexio builds advertising data pipelines (ETL), a campaign builder and AI agents for advertising teams, and delivers custom AI agents, MCP servers and cloud software. The answer most people ask for first: yes, Vexio works end to end on Amazon, Walmart, Google, Meta, TikTok and Shopify, as a service. Each topic below links to the page that answers in full.
By The Vexio team Updated
About Vexio 8 answers
- What is AI-first advertising automation?
- It means building advertising workflows around AI from the start, with a person approving anything that changes an account. Vexio works in two areas. Ad data and automation covers advertising data pipelines (ETL), the campaign builder and an AI advertising agent. Custom AI and software covers AI agents, MCP servers and cloud software built to order.
- What is marketing ETL?
- Marketing ETL means extracting data from advertising sources such as Amazon Ads, Google Ads and Meta Ads, transforming it into clean, consistent tables and loading it into a warehouse you control. Vexio builds these pipelines as a service, scoped per engagement. See advertising data pipelines (ETL).
- Which ad platforms does Vexio work on?
- Amazon Ads, Walmart Connect, Google Ads, Meta Ads, TikTok Ads and Shopify. On each one we deliver end-to-end work as a service: account structure and campaign build, data pipelines and reporting, automation, agents and MCP, and ongoing optimisation. The platforms page covers each one.
- What is the difference between the campaign builder and the AI agent?
- The campaign builder starts from a file: you supply a spreadsheet of campaigns, it checks every row, and you approve before anything is published. The AI advertising agent starts from a request in plain language: it plans the steps, calls tools through MCP and returns changes for your approval. Both sit on the agents page. See the campaign builder and the AI advertising agent.
- Do you build custom AI agents and MCP servers for other companies?
- Yes. An engagement starts with a scoping conversation about the job the agent should do, the tools it may call and the data it may read. Then come a prototype, a build with evaluation and approval steps, and a handover. Details are on the custom AI agent development and MCP server development pages.
- What technologies do you work with?
- Node.js, Python and Django for application code, BigQuery and PostgreSQL as warehouse destinations, the Model Context Protocol (MCP) for connecting AI agents to tools, and Cloudflare, AWS and Azure for hosting, with DevOps practice around them: deploy pipelines, infrastructure as code and monitoring. See custom software and cloud and DevOps.
- Is Vexio a software product or a service company?
- Both, depending on the work. Advertising data pipelines, custom AI agents, MCP servers, software and cloud work are services: we scope, build and hand over per engagement. The campaign builder and the AI advertising agent are Vexio's own tools for advertising teams. On the ad platforms themselves we also deliver end-to-end service work.
- How do I book a demo or start a project?
- Send a request on the demo page with your name, work email, company and what you want to automate or build. Mention the platforms and data involved if you can. We reply by email to arrange a demo or a scoping call.
Advertising data pipelines (ETL) 8 answers
Advertising data pipelines (ETL)
Marketing ETL, BigQuery loads, backfills and how an engagement runs.
Read the page →- What is marketing ETL?
- Marketing ETL is the process of extracting data from advertising and commerce sources, transforming it into clean, consistent tables and loading it into a warehouse. It turns the separate reports of Amazon Ads, Google Ads, Meta Ads and others into one dataset you can query. Vexio builds these advertising data pipelines for you as a service.
- How do you get Amazon Ads data into BigQuery?
- We build a pipeline that requests reports through the Amazon Ads API, polls until each report is ready, downloads it, cleans and merges it, and loads it into BigQuery tables defined in configuration. Credential review, schema design and a backfill of history are part of the engagement. Where near-real-time data matters, we discuss Amazon Marketing Stream while scoping. For other sources, see Google Ads to BigQuery options.
- What is the difference between ETL and ELT for advertising data?
- In ETL the data is transformed before it is loaded; in ELT it is loaded raw and transformed inside the warehouse. For advertising data the practical answer is often both: raw files land in storage as a record of what the source returned, and cleaned tables are loaded for analysis. We recommend one design during scoping, with reasons. More in marketing ETL explained.
- How long does a data pipeline project take?
- It depends on the number of sources, how much history you need and the access you already have, so we do not quote a standard duration. We scope it on a call and send a written scope that lists the sources, the destination, the history and the milestones before any work starts.
- Which warehouses and clouds can you load into?
- BigQuery and PostgreSQL, with raw files landed as parquet in Amazon S3, Google Cloud Storage or Azure Blob. The destination is in your own account, so your data stays under your control. An AWS data pipeline for ad data is part of this work, and hosting runs on AWS, Azure or Cloudflare as agreed. See cloud and DevOps.
- What happens when a source fails or data arrives late?
- Failed calls are retried with increasing delays, and loads are idempotent, so a re-run does not duplicate rows. A ledger of loaded dates flags any date that is missing or stale, and late-arriving data can be backfilled. Alerts go to Slack or email, so a person hears about a gap before a report shows it.
- Can you pull Amazon SP-API, Walmart Connect or TikTok Ads data?
- Yes, as sources we build connectors for: the Amazon Selling Partner API for sales and orders, Walmart Connect and TikTok Ads for advertising reports, alongside Amazon Ads, Google Ads and Meta Ads. Each connector is built against the access you have. See what we do on each platform, or ask about a source that is not listed.
- Is this a product I can buy?
- No. Advertising data pipelines are a service: we build connectors for the sources you need, deploy them with infrastructure as code and hand them over. A source on our list means we build a connector for it in your engagement, not that a ready-made connector can be switched on.
Campaign builder 8 answers
- What is campaign management software with bulk upload?
- It lets an ad team create or change many campaigns from one file instead of one at a time in each ad platform. The Vexio campaign builder reads the file, maps its columns to campaign fields with AI, validates every row, sorts rows into ready, needs input and error, and publishes only after you approve.
- Is this a Google Ads Editor alternative?
- Partly. Google Ads Editor is Google's free desktop application for editing an account in bulk, and for most single-account edits it is the better tool. The campaign builder is aimed at a different job: validating a campaign file row by row, answering open questions once and approving before publishing. See Google Ads Editor vs bulk upload vs API.
- How does a Google Ads bulk upload differ from the Vexio campaign builder?
- A Google Ads bulk upload applies the file you give it and reports the problems it finds. The Vexio campaign builder is designed to check the file before anything is sent, sort rows into ready, needs input and error, and ask you about ambiguous columns. You then approve what goes out.
- Can you help with Amazon bulk operations or a Meta Ads bulk upload?
- Yes, as a service. Vexio prepares and checks the files for Amazon Ads bulk operations spreadsheets and Meta Ads spreadsheet imports, and loads the campaigns. See Amazon Ads and Meta Ads for what we do on each, or describe the job on the demo form and we will say how we would run it.
- Do you offer Amazon PPC software or Amazon Ads management?
- Vexio does not sell Amazon PPC software. It delivers end-to-end Amazon Ads work as a service: account structure, campaign build, reporting and optimisation. See the Amazon Ads page. If you need Amazon Ads data in a warehouse, see marketing ETL.
- How does the campaign builder know which column is which?
- It analyses the column headers and the values beneath them and proposes a campaign field for each column. Clear columns are mapped automatically. Ambiguous columns, or columns that pack several values into one cell, are flagged for you to confirm rather than guessed.
- Does anything go live without my approval?
- No. After validation you review the result table, answer any open questions and approve, and only approved rows are published. Which actions need approval and how the review screen behaves are shown in the demo, so ask to see them before you decide.
- What happens to rows with errors?
- They are listed separately with the reason, so you can fix the file or correct the row in the table. Rows that are ready can move forward while the rest wait, and rows that need input are held until you answer the question they raise.
Ad platforms we work on 5 answers
Ad platforms we work on
What Vexio does on Amazon, Walmart, Google, Meta, TikTok and Shopify.
Read the page →- Which ad platforms does Vexio work on?
- Vexio works end to end on Amazon Ads, Walmart Connect, Google Ads, Meta Ads, TikTok Ads and Shopify. On each that means account structure and campaign build, data pipelines and reporting, automation, agents and MCP, and ongoing optimisation support. Tell us which platforms you use and we scope just that.
- What does an end-to-end platform engagement include?
- It is some mix of five areas: account structure and campaign build, data pipelines and reporting, automation, agents and MCP, and ongoing optimisation support. We agree which ones you need in a scoping conversation. Some teams only want the reporting pipeline, and that is fine.
- Can you put Amazon, Walmart, Google and Meta data in one warehouse?
- Yes. We build a connector for each source and load them into BigQuery or PostgreSQL in your own account, with the same definitions for spend, sales and ratios across platforms. Attribution windows still differ between platforms, and we document those differences. See advertising data pipelines (ETL).
- Do you build campaigns in bulk for these platforms?
- Yes. Building campaigns from a spreadsheet, and checking every row before it is uploaded, is part of the campaign build work on each platform. Each file is checked for missing or inconsistent values first. See the campaign builder for how bulk campaign upload works.
- Is Vexio an official partner of these platforms?
- No. Platform names are trademarks of their owners and appear as text only. Vexio is an independent company and is not affiliated with, endorsed by or certified by Amazon, Walmart, Google, Meta, TikTok or Shopify. Their own terms apply when you use their services.
AI advertising agent 7 answers
- What is an AI ads manager, and what is the Vexio AI advertising agent?
- An AI ads manager is software that helps run advertising accounts using an AI model. The Vexio AI advertising agent is built for advertising teams. You describe a task in plain language, such as reviewing campaigns or preparing a change. The agent plans the steps, uses tools through MCP to read data and draft changes, and presents the result for review.
- What is an AI agent, compared with a chatbot?
- A chatbot answers a message. An AI agent pursues a goal: it breaks the goal into steps, calls tools such as an ad platform API, checks the results and continues until the work is done or it needs to ask you. For definitions, see the glossary.
- Is this AI PPC management? Why does the agent use MCP?
- It is aimed at the work behind AI PPC management: reading performance, drafting changes and waiting for approval. It uses MCP, the Model Context Protocol, an open protocol for connecting AI applications to tools. Each action is an MCP tool, so one can be added, restricted or removed without rewriting the agent. See what an MCP server is and MCP server development.
- Will the agent change my campaigns without asking?
- No. The agent is designed to propose changes and wait for approval before anything is published, and every request, tool call and outcome is logged. Which actions need approval, and which limits apply, are agreed with you and shown in the demo.
- Can an AI ads manager replace a PPC manager?
- It is built to take over repetitive work, such as pulling performance, spotting changes and drafting edits, while a person decides and approves. It does not set strategy or budgets on its own. For ACOS and TACOS terms an agent reports on, see ACOS vs TACOS.
- Does it work with Amazon Ads, Google Ads or Meta Ads?
- Vexio does end-to-end work on Amazon Ads, Walmart Connect, Google Ads, Meta Ads, TikTok Ads and Shopify as a service, described on the platforms page. For the agent itself, tell us your platform on the demo form and we will say what the agent can do there.
- Can I get an agent built for my own workflow instead?
- Yes. That is the custom AI agent development service: an agent for your tools, data and approval rules, built and deployed for you. The first step is a scoping conversation about the job, the tools and who approves the result.
AI agents hub 6 answers
- What are AI agents for advertising?
- They are AI agents that do work around advertising and operations: preparing campaigns, answering support questions, assembling reports, researching competitors and moving information between tools. Each has a defined job, a limited set of tools and a person who approves what it proposes.
- Is the campaign builder an AI agent?
- It is one tool among several. The campaign builder turns a spreadsheet into validated, approved campaigns, using AI to map columns. The AI advertising agent takes requests in plain language. The other tiles on this page are different jobs, each with its own scope.
- Do agents act without approval?
- The pattern we build around is propose, review, approve, execute, with an audit log. Which actions need approval is decided with you for each engagement, and anything that changes a live ad account or system waits for a person by default. Every request and tool call is recorded in the audit log.
- Do you offer AI agent development for other companies?
- Yes. Custom agents are delivered as engagements: a scoping conversation, a prototype on a sample of your data, a build with agreed test cases, and a handover. See AI agent development, or read how to choose an AI agent development company.
- Is Vexio a marketing automation agency?
- Vexio is an engineering company. We build the systems behind marketing automation: advertising data pipelines, campaign tooling and AI agents. We also deliver hands-on work on Amazon Ads, Walmart Connect, Google Ads, Meta Ads, TikTok Ads and Shopify as a service. See the platforms we work on.
- How do I start a custom agent project?
- Describe the job on the demo form: what the agent should do, which tools it would use and who approves the result. We reply by email to arrange a scoping conversation, and you receive a written scope before work starts.
Custom AI agents 8 answers
- What does an AI agent development company do?
- An AI agent development company designs, builds and deploys AI agents for other organisations. Vexio scopes the job and its limits, builds a prototype on a sample of your data, then builds the agent with the tools it may call. We test it against cases you agree, add approval and logging, deploy it in your cloud and hand it over with documentation.
- How much does custom AI agent development cost?
- We do not publish a price list. Cost depends on how many tools the agent calls, how clean your data is and which actions need approval. We scope the job in a first conversation and quote against that written scope. The article on custom AI agent development process and cost factors explains what moves the number.
- How long does an AI agent development project take?
- It depends on scope, so we do not quote a standard duration. A short prototype comes first, built on a sample of your data, so you can judge the result before you commit to the full build. The scoping call gives you a timeline for your case.
- Which AI models do you use for custom AI agents?
- The model is chosen per project, based on the task, your data-handling constraints and the cloud you run on. We design the agent so the model can be changed later without rewriting the tools around it. Tell us about constraints such as data residency or an approved provider in the scoping call.
- How do you stop an AI agent doing something wrong?
- By limiting what it can do, not by hoping it behaves. Each tool gets the smallest permission that works, write actions can require human approval, every call is logged, and the agent is tested against agreed cases before and after each change.
- Do you need access to our production data?
- Not at the start. A prototype can run on a sample or on redacted data. If production access is needed later, it is agreed in writing with the minimum scope the agent needs to do its job. We can also work inside your environment instead of receiving a copy of the data.
- Who owns the AI agent you build?
- Ownership and licensing terms are written into the engagement agreement before work starts, so ask about them on the first call. The website does not set those terms. Custom projects run under a separate written agreement. If you want something specific, such as owning the code outright, say so early so the scope reflects it.
- How do I choose an AI agent development company?
- Ask for a prototype on your own data, a written scope, evaluation cases you agree and the approval rules for any action that changes something. Be wary of a vendor who cannot say what the agent may not do. Our guide on how to choose an AI agent development company lists the questions.
MCP servers 7 answers
MCP servers
What MCP is, how it differs from a REST API and what building a server involves.
Read the page →- What is an MCP server?
- An MCP server is a program that exposes tools, data and prompts to AI applications through the Model Context Protocol, an open protocol for connecting AI applications to external systems. An AI client connects to the server, lists what it offers and calls the tools. See what is an MCP server.
- How is MCP different from a REST API?
- A REST API is written for developers, who read the documentation and write code against it. An MCP server describes its tools in a form an AI application can discover and call itself. Many MCP servers sit on top of an existing API: the API stays, and the server adds the descriptions, permissions and approval rules an agent needs. Read MCP vs REST API.
- Can you build an MCP server for our internal API or database?
- Yes. MCP server development is a service we deliver. A typical project wraps an existing API or database with a small set of well-named tools, read-only first, with authentication, rate limits and logging. You get the server, its documentation and a handover session.
- How do you build an MCP server safely?
- The engineering pattern is a dry-run default for anything that writes, an audit log of every call and a scoped identity for each caller, so the server can do only what that caller is allowed to do. We also validate every input and test with hostile input before launch. Details are agreed per project.
- Should an MCP server run locally or remotely?
- A local server runs on the user's machine and talks to the client over standard input and output. A remote server runs on a host you control and is reached over HTTP, which suits shared tools and central access control. We build either, and one codebase can serve both.
- Is it safe to let an AI call our systems?
- It is as safe as the permissions you give it. We scope credentials to the minimum, make write actions explicit and approvable, log every call and test the tools against abuse cases such as unexpected input. Treat an MCP server as an API that an unpredictable client will call.
- How do I build an MCP server in Node.js or Python?
- An MCP server is a normal program that uses an MCP SDK to declare tools with typed inputs and answer calls from a client. For a step-by-step example, read how to build an MCP server in TypeScript. If you want it built, hosted and handed over, that is the service Vexio delivers.
Custom software 7 answers
Custom software
Node.js, Python and Django projects: scope, stack, hosting and handover.
Read the page →- What does custom software development include?
- We build web applications, internal tools and back-offices, APIs and integrations, and data pipelines, mostly in Node.js and Python, with Django where a full framework fits. You get the code, tests, documentation and a repeatable deployment. The same team also builds custom AI agents, MCP servers and advertising data pipelines.
- Node.js, Python or Django: which should we use?
- The problem decides. Node.js suits APIs and real-time services, Python suits data and AI work, and Django suits applications that need an admin area, sign-in and a relational database from day one. Before the scope is agreed, we tell you which one fits your project and why.
- Can you take over or extend our existing codebase?
- Yes. We can take over, extend or modernise an existing application. The first step is a review of the code, the deployment and the tests, which tells us what is safe to change and what needs work before anything new is added.
- How does a custom software project run?
- In short cycles, with working software at the end of each one that you review. Scope, milestones and how we communicate are written down at the start. At each review you see the software running, not a slide deck. The project ends with deployment, documentation and a handover session.
- Where is the software hosted?
- On Cloudflare, AWS or Azure, in your own account if you prefer. We set up the hosting, the deployment pipeline and monitoring as part of the project. We can also deploy to a setup you already run, if that is where your team works. See cloud and DevOps for what that work includes.
- How much does custom software development cost?
- We do not publish a price list. Cost depends on scope, integrations and how much needs to be built from scratch. We agree a written scope after a first conversation and quote against it. Describe the project on the demo form and we reply by email.
- Can you add an AI agent or an MCP server to our software?
- Yes. If an application should let an AI assistant read data or take actions, we add an MCP server with scoped permissions and approvals, or build an agent around one job. Write actions can wait for a person to approve them. See AI agent development and MCP server development.
Cloud and DevOps 6 answers
- What does DevOps consulting include?
- We design and set up hosting on Cloudflare, AWS or Azure, build and deploy pipelines, infrastructure defined as code, and monitoring and alerting, then write the runbooks to operate it. It can be a new setup, a migration or a review of what you run now.
- Can you build an AWS data pipeline?
- Yes. We build data pipelines on AWS: raw files land in Amazon S3 and cleaned tables load into your warehouse. The design is described on the advertising data pipelines page, and marketing ETL explained covers the concepts. The same pipelines can run on Google Cloud or Azure when your data lives there.
- Cloudflare, AWS or Azure: which should we use?
- It depends on what you already run and what you need. Cloudflare Workers suit light APIs and sites close to users. AWS and Azure suit larger or regulated workloads and teams that already have an account and skills there. We give a recommendation with reasons and trade-offs, not a default.
- Do you host and operate things for us?
- By default we set up systems in your accounts and hand them over. Ongoing operation is agreed per engagement, so say on the first call if you want us to keep running it. Common arrangements are a handover with a runbook, or a support agreement after launch.
- Can you review our existing cloud setup?
- Yes. A review looks at deployment, access, backups, monitoring and cost in what you run now, and ends with a prioritised list of changes. You can act on the list yourselves or ask us to do the work. It is a sensible first step before a migration or a new data pipeline.
- Do you do Cloudflare Workers development and AI hosting?
- Yes. Cloudflare Workers development is part of the same service, and so is hosting for AI agents and MCP servers. Workers suit light APIs, sites and remote MCP servers that must be reachable over HTTP. See custom AI agents and MCP servers.
Talk to us Demo or question
Not answered here?
Ask us directly. We reply by email, and we add good questions to this page.
- hello@vexio.ai
- Phone
- +1 778 896 4353
- Message us
- Web
- vexio.ai