AI for Owner-Run Businesses | Ignitara Blog

Can AI Work With the CRM I Already Have?

Justin Farmer — Wed, 28 Oct 2026 11:46:00 GMT

Yes. In almost every case, AI can work with the CRM you already have, and you should insist on it. If someone tells you that using AI means switching to their platform, that's a sales pitch, not a technical requirement.

This is the question I hear most often on assessment calls right now, usually with a tired edge to it. Owners have been sold a new system before. They've migrated data, retrained a team, and watched the shiny thing become the thing nobody opens. So when AI comes up, the first question isn't "what can it do," it's "do I have to start over?"

You don't. Here's how it actually works, and the two things to check before you believe anyone, including me.

How does AI connect to a CRM?

Think of your CRM as a filing cabinet and the AI as a very fast assistant. The question is whether the assistant has a key to the cabinet.

Most CRMs built in the last decade have a way to hand out that key. The technical term is an API, or more recently an MCP connector. You don't need to know how either works. You need to know whether your CRM offers one, and most do. Salesforce and everything built on it, HubSpot, GoHighLevel, Zoho, Pipedrive, and the mortgage- and real-estate-specific CRMs that sit on top of those platforms all have one.

Once the connection exists, the AI can do the things you'd ask an assistant to do with the filing cabinet: find the contacts who match a description, summarize what's happened with a client, draft the follow-up, and, if you let it, write the note back when you're done.

Nothing moves. Your data stays where it is. The AI reads from it and writes to it in place.

What's the difference between read access and write access?

This is the part that separates an AI setup that works for a day from one that works for a year.

Read access means the AI can look at your CRM. That's enough for a lot of useful things: "who haven't I talked to in 60 days," "summarize this client's history before my call," "which deals are stuck."

Write access means the AI can also update your CRM. Log the call. Change the stage. Add the note. Update the last-contact date.

Most people stop at read access, because it's the default and it feels safer. Then they're surprised when the AI gives them the same list every morning. That's what happened to a loan officer I work with, and I wrote up the whole story here: why his AI call list kept repeating. The short version is that without write access, the AI has no memory of what you did yesterday, so it starts from scratch every day.

If you want AI to take a recurring task off your plate, it needs to be able to close the loop. That means write access, with limits.

What limits should I put on it?

Three, and they take about five minutes.

No delete. Turn on create and update. Leave delete off. There is no everyday task where an AI needs to delete a record in your CRM, and leaving it off means the worst case is a bad note, not a lost client.

Approval before writes. The AI tools I use show you what they're about to change and wait for a yes. Keep that on, at least for the first month. You'll see exactly what's landing in your CRM and catch mistakes before they pile up.

Your account, not theirs. The connection should be set up inside your own CRM and your own AI account, with your credentials. If a vendor wants to route your CRM through their system, ask why. The answer is usually that it keeps you dependent on them.

Do I need a new tool for any of this?

Probably not. You likely already pay for one of the big AI assistants, or your company does. Claude, ChatGPT, and Microsoft Copilot can all connect to CRMs. Which one you use depends mostly on what your business already runs on. If your email and calendar live in Microsoft, Copilot is often the path of least resistance because the permissions are already there. If not, Claude or ChatGPT with a connector to your CRM is the usual answer.

The tool matters less than the setup. I've seen the same workflow succeed and fail on the same tool, and the difference was always whether the connection was done right.

What should I try first?

Pick one list you build by hand every week. Who to call, who to follow up with, which deals need a nudge. Something you do from memory or from scrolling through the CRM.

Then try this:

Connect your AI assistant to your CRM with read access only.
Ask it to build that list, with a sentence of context on each person.
Check the list against what you'd have picked yourself.

If the list is close, you've found your first project. Add write access with the limits above, teach the AI to log what you did, and you have a system that gets a little smarter every day instead of resetting every morning.

If the list is way off, that's useful too. It usually means the CRM data is thinner than you thought, and that's a cheaper problem to find now than after you've built on top of it.

Questions owners ask

Will connecting AI to my CRM put my client data at risk?

Not if it's set up in your own accounts with limited permissions. The AI reads and writes in place; nothing is copied out. Keep delete off, keep approval on, and the exposure is smaller than giving a new employee a login.

My CRM is old. Can AI still work with it?

Maybe. If it has an API or an export, there's usually a path. If it has neither, the CRM is the bottleneck, and that's a different conversation than "which AI."

Can AI replace my CRM?

No, and you wouldn't want it to. The CRM is where the record lives. The AI is what makes the record useful. You need both.

How long does the connection take to set up?

Usually under an hour, including the permission puzzles. The first time through, it's worth doing with someone who has done it before, because the failure modes are quiet and the fix is often one setting.

Want to see if your CRM is ready?

I'm Justin Farmer, and I run Ignitara out of Raleigh, NC. I help owners put AI to work inside the tools they already pay for, and we build it together so they own it.

If you want to know whether your CRM can connect, or you've tried and it didn't stick, get your free AI Plan. Bring the name of your CRM. I can usually tell you on the call.

Why a Loan Officer's AI Call List Keeps Showing the Same People Every Day, and the One Setting That Fixed It

Justin Farmer — Wed, 28 Oct 2026 11:45:00 GMT

An AI call list keeps showing the same people every day because the AI has no memory of who was already called, and it could not update the CRM. The fix wasn't a new tool, it was a permission setting in the CRM he already paid for.

A loan officer I work with built his own AI call list before we ever met. It ran every morning at 8, pulled realtors out of his CRM, and told him who to call and why. Smart setup. It had one problem: it kept handing him the same names.

This is the story of why that happened, what we did about it in two working sessions, and what it says about where AI actually breaks down for a business owner. Short version: the AI was fine. It just had no memory, because nobody had given it one.

What was he trying to do?

He's a mortgage loan officer in the Triangle. A big share of his business comes from realtor referrals, so staying in front of realtors matters more than almost anything else on his calendar. His plan was simple: call 20 or so realtors every Monday, split between the ones who just sent him a deal (say thank you) and the ones who have gone quiet (re-engage).

He set up two scheduled tasks in Claude himself, a few weeks before our first session. Each morning, one task pulled a list of past clients to call and the other pulled realtors, with suggested talking points for each. He was ahead of most owners I meet. He had already figured out what AI should do for him and built a first version on his own.

Then he opened it on day three and saw the same three realtors from day one. And day two.

In his words: "It's giving me the same realtors to call almost every day."

Why did the list keep repeating?

Here's the part most people miss. The AI wasn't broken. It was doing exactly what it was asked: look at the CRM, find realtors who haven't been contacted in a while, list them.

The problem was that nothing told the CRM he had made the call. His connection to the CRM was read-only, so Claude could look but not write. After he called a realtor, the "last contact" date never changed. The next morning, that realtor still looked untouched, so back on the list they went.

He also admitted what most of us would admit about logging calls: "I don't do a really good job at that." His team's convention was to add a line to each contact's notes after every touch, with initials in front. Good practice. Easy to skip when you have 20 calls to make and a closing at 2.

So the real bottleneck wasn't the call list. It was the logging. The list was just where the missing logs showed up.

What did we build in session one?

Two things, in this order.

First, we gave the AI permission to write to his CRM. His CRM is built on Salesforce, which has a setting for exactly this. We turned on create and update, and left delete off. There is no scenario where an AI should be deleting records in your CRM, so we didn't give it the option.

This took a couple of tries. The first connection we set up wouldn't accept the write permission. The second one needed a credential we had to go find in a different settings page. That's normal. Connecting AI to real business software is mostly a series of small permission puzzles, and the reason I sit on the screen with owners instead of sending instructions.

Second, we turned his morning prompt into a skill. A skill is just a written set of instructions the AI follows the same way every time. His has two jobs:

Build the list. Pick 10 realtors (he can change the number), sorted into thank-you calls and re-engage calls, with talking points.
Log the call. When he says "I talked to so-and-so, here's what we discussed," the AI writes the note into the contact's record with his initials in front, the way his team already does it, and updates the last-contact date. That realtor then drops off the list for 21 days.

We tested it on one real contact. The note showed up in the CRM, initials and all. He laughed at the initials. Then the morning task was swapped to a one-line prompt that just runs the skill.

What went wrong between sessions?

Two weeks later, the list was still repeating.

Not because the skill failed. Because logging through it hadn't become a habit yet, and a couple of edits to the skill in the meantime had pointed the notes at the wrong field. When we logged three calls together, the AI said "notes added," and the notes weren't there.

We found it in about 10 minutes. Logged one contact at a time, confirmed the right field, watched for the approval prompt the AI shows before it writes anything. The notes landed. "This is exactly what we wanted to do. Look at that."

I'm including this part on purpose. The first build is rarely the last build. The second session is where it starts to stick, because now you're fixing something you've actually used.

What happened next?

With the call list working, we had time left in the session, so I showed him how I prep for my own meetings. I have a skill that pulls the calendar invite, the email thread, and anything I know about the person into a short brief. He watched it run and asked, "What did you just do to get all of that that fast?"

So we built him one. His company runs on Microsoft, so we built it in Copilot instead of Claude, using the tools he already has. It runs at 5 a.m., reads his calendar and email, and writes a brief for each meeting that day. We tested it on a real coffee meeting he had that week. It caught a detail from the email thread that he had forgotten. "That's pretty wild, isn't it?"

That brief now goes out to two teammates too. It still needs him to click send, which is a company setting, and that's the next thing on the list.

What's the lesson for your business?

If you've tried AI and it felt dumb, check whether it can remember anything.

Most AI tools, out of the box, start from zero every time. They can read your data and give you a smart answer, but unless you connect them to something that keeps state, like your CRM, they have no idea what happened yesterday. That's not a flaw in the AI. It's a setup step that almost nobody does.

The loan officer's first version was 80% right. The missing 20% was permission to write back. Once the AI could close the loop, the whole thing worked. And the fix wasn't a new tool. It was a setting in the tool he already paid for.

Questions owners ask

Can AI work with the CRM I already have?

Usually, yes. Most modern CRMs, including anything built on Salesforce, HubSpot, or GoHighLevel, can connect to AI tools for reading and writing. The write part is the one people skip, and it's the one that makes the AI useful day after day.

Why does my AI call list keep showing the same people every day?

Because the AI has no memory of who you already called, and it could not update the CRM. Connect it to your CRM and give it permission to update it.

Is it safe to let AI write to my CRM?

Yes, with limits. Turn on create and update, keep delete off, and use a tool that asks for approval before it writes. You should see every change before it happens.

Do I need to know how to code to set this up?

No. The loan officer in this story isn't a programmer. The setup is permissions and plain-language instructions, and we do it together on his screen.

Want to see where this fits in your week?

I'm Justin Farmer, and I run Ignitara out of Raleigh, NC. I help owners put AI to work inside the tools they already use, and we build it together so they own it.

If you have a task that keeps repeating, or an AI setup that never quite stuck, get your free AI Plan. We'll find the one thing worth fixing first.

Where Should a Small Business Start With AI?

Justin Farmer — Sat, 10 Oct 2026 14:28:23 GMT

Start with whatever is standing between you and the work only you can do. Not with a tool, not with a course, and not with the flashiest thing you saw online this week. Find the one task that keeps pulling you away from your best work, and hand AI that first.

That's my whole answer. The rest of this post is why I believe it, and how you can find that task in your own business without hiring anyone.

Why do most owners start with AI in the wrong place?

Most owners I talk to start with the tool. They hear about a chatbot, an AI phone answering service, or a new app, and they try to find a spot in the business to plug it in.

I get it. The tools are exciting, and everyone is selling one. But starting with the tool is backwards. It's like buying a nail gun and then walking around the house looking for something to nail.

When you start with the tool, you end up automating whatever the tool happens to be good at. That's rarely the thing that's actually costing you. So the tool sits there, you pay for it every month, and nothing about your week really changes.

The better starting point is your own week. What does it look like right now, and what keeps getting in the way?

What did my last few assessment calls teach me?

On every free assessment call, I ask the owner a simple question: what is an hour of your time worth?

I ask because the answer usually tells me what's worth taking off their plate. If your hour is worth a lot, then every hour you spend on low-value work is expensive, and that's where AI should go first.

3 of my last 4 assessment calls gave me the same surprise. Nobody answered in hours.

One owner told me saving an hour wouldn't change a thing for him. Saving a day would.

Another said his only real limit is that he can't be in 2 places at once.

The third said he's at his best face to face with his clients, and everything else just gets in the way.

I sat with that for a while after the calls. None of these owners wanted AI doing their best work. Not one of them asked me to automate the part of the job they're great at. They wanted AI clearing the path to it.

That changed how I think about the hour question. It still works. It just points somewhere new. Instead of asking "what's an hour worth," the better follow-up is "what's in the way of the work only you can do?" That's where the first AI project belongs.

How do I figure out what to hand to AI first?

You don't need me for this part. You can do it yourself this week with two questions and a piece of paper.

The first question is the one above: what's an hour of your time worth? Don't overthink the number. The point isn't the math. The point is to notice when you're spending that hour on something that isn't worth it.

The second question is one I ask on every call too: what would you pay the most to make disappear?

Not "what could AI do." What do you dread? What do you keep putting off? What makes you sigh when it shows up on your calendar or in your inbox?

Write down your answers to both. Then look at your list and ask one more thing about each item: does this pull me away from the work only I can do?

The item that does that the most is your starting point. It's usually not glamorous. It might be the same five answers you type out over and over. It might be the follow-ups you mean to send and never do. It might be the paperwork that piles up after every client meeting. That's fine. Boring is good. Boring is where AI saves the most time with the least risk.

Do I need a big AI system to get started?

No. And I'd steer you away from one at the start.

The owners I work with almost never need a big system first. They need one small thing that works, that they actually use, and that gives them back time they can feel. Once that's in place, the next thing gets easier, because now you know what good looks like in your own business.

A big system built before you know your real bottleneck is a guess. A small fix aimed at your real bottleneck is a decision.

What should AI never do in my business?

This is the part people skip. Before you hand anything to AI, decide what you'd never hand to it.

For the 3 owners I mentioned, the answers pointed the same way. One wanted whole days back, not stray hours. One wanted to stop being stuck in 1 place. One wanted more time face to face with his clients. Each of them was describing the work he built the business on, and that's the work AI should protect, not replace.

Your list might look different. Maybe it's the way you quote a job, or how you talk to a customer who's upset. Whatever it is, write it down. It keeps you honest when someone tries to sell you a tool that promises to "do it all."

The goal isn't to remove you from your business. It's to give you more of the parts of your business that only you can do.

Questions owners ask

What's the first thing a small business should use AI for?

The task that most often pulls you away from your best work. For a lot of owners that's repeat questions, follow-ups, or admin that piles up after meetings. Start with one, not five.

Do I need to be good with technology to start using AI?

No. The hard part isn't the technology. It's knowing which task is worth handing off. If you can describe what eats your week, you can start.

How do I know if AI is actually helping?

Ask yourself if you got time back for the work you care about. If the tool saves time on paper but you're still stuck doing the same things, it's aimed at the wrong problem.

Should I hire someone or figure out AI on my own?

Try the two questions in this post on your own first. If you find the task but aren't sure how to hand it off safely, that's when a second set of eyes helps.

Want help finding your starting point?

I'm Justin Farmer, and I run Ignitara out of Raleigh, NC. I help owners figure out where AI actually fits in their business, then we build it together in their own accounts.

If you'd like to talk through your own week, get your free AI Plan. We'll look at what's standing between you and your best work, and talk about where AI could help first.