Insights
Practical answers for marketing leaders.
Guidance on marketing technology, measurement, websites and search, written by practitioners and useful whether or not you ever hire us.
Chat, skills, agents and coding agents: what's the difference?
Four ways to put AI to work, from a conversation you drive to software that acts on its own. Here's what each one is, what it's good for, and how to choose.
Read the article- Chat is a conversation you drive. It's the best place to start, and it forgets the job when the chat ends.
- A skill packages instructions, reference files and scripts so the AI does one task the same way every time.
- An agent works toward a goal, uses your tools and takes actions, with people approving what matters.
New series
AI 101: a practical introduction to AI for marketing teams
Six short, practical guides that take a marketing team from its first chat to knowing when an agent is worth building. Read them in order, or jump to what you need.
- 01Chat, skills, agents and coding agents: what's the difference?
- 02How AI language models work: a plain-English guide for marketers
- 03How to write AI prompts that work: a practical guide for marketing teams
- 04How to choose AI tools for your marketing team
- 05An AI usage policy for marketing teams: what to include
- 06Your first 30 days with AI: a practical plan for marketing teams
Your first 30 days with AI: a practical plan for marketing teams
A marketing team can get real value from AI in a month without a big program. Pick a few tasks, measure how long they take today, try AI on real work, keep what helps and drop what doesn't.
Read the articleHow to write AI prompts that work: a practical guide for marketing teams
Good prompts are mostly good briefs. Here is a framework your team can use today, templates for common marketing tasks, and a sense of when a prompt should become something you save.
Read the articleHow to choose AI tools for your marketing team
New AI tools launch every week, and most marketing teams already pay for more AI than they use. The way through is to start from the work, test candidates on your own examples, and buy only what earns a place.
Read the articleHow AI language models work: a plain-English guide for marketers
What is actually happening when you type into ChatGPT, Claude or Gemini, why it sometimes gets things wrong, and what that means for how your team should use it.
Read the articleAn AI usage policy for marketing teams: what to include
Your team is already using AI, with or without permission. A one-page policy with clear rules does more to protect the brand than a ban nobody follows. Here is what to put in it, plus a template you can copy.
Read the articleWhat does a marketing technology consultant do?
A plain description of the job and the documents it should leave behind, plus a few ways to spot a reseller dressed up as a consultant.
Read the articleWebsite redesign vs. optimization: how to decide
If the platform or the site structure is what's holding you back, a rebuild is probably justified. If a handful of pages are underperforming, fix those pages. This article helps you work out which situation you're in before the budget is committed, and how to keep your search traffic if you do rebuild.
Read the articleMarTech consultancy vs. marketing agency: which do you need?
Agencies and MarTech consultancies solve different problems. Work out which problem you actually have before you hire anyone.
Read the articleHow to measure digital marketing performance (a framework for leadership teams)
A leadership team needs a handful of metrics tied to business objectives, someone accountable for the data behind each one, and a realistic sense of what attribution can and can't prove. Forty KPIs on a slide won't do that job.
Read the articleHow to choose a MarTech stack (without buying tools you won't use)
Most stacks grow one urgent purchase at a time. A better way starts with the work your team has to do and the data that needs to move, and treats every vendor as a candidate until it earns its place.
Read the articleHow much does MarTech consulting cost? What drives the price
Nobody can give you an honest single number. What we can give you is the list of things that set the price, and you control more of them than you'd expect.
Read the articleGA4 and Google Tag Manager implementation best practices
When GA4 numbers can't be trusted, the tool is rarely the cause. The usual culprits are a missing measurement plan, event names that drifted and a tag container nobody owns. This guide covers the setup that gets numbers through a leadership meeting intact.
Read the article
Next step
Have a question we haven’t answered?
Ask us directly. If it’s a common question, it may become the next article.
Prefer email? contact@atlmartech.com