From Resistance to 90% Adoption: How To Build an AI Legal Front Door Lawyers Actually Use

At some point or another, you've probably heard "legal is the holdup." It's something that gets casually thrown around in meetings, but if you're sitting on the legal side of the table, it's hard not to take personally — especially when you know exactly how much work is actually going in.

But, another reason why this isn’t nice to hear is that lawyers know it's at least partly true. Requests fall into email inboxes and don't resurface until someone chases them, work gets sent to outside counsel because it's easier than figuring out who internally has capacity, and deadlines sometimes slip because no one could see where the bottleneck was in the first place.

The response, almost universally, is to go looking for technology. A new tool. A smarter system. Something with AI in the name. And so the procurement process begins with demos, shortlists, and business cases, and eventually something gets implemented. 

However, many teams overlook the fact that you can implement the best legal tech solutions in the industry, but if adoption doesn’t follow, nothing will change.

That was the reality Joy Thorpe, Director of Strategy for the Legal Centre of Excellence at Altria, walked into when her department was tasked with modernizing. The mandate was to be quicker, more efficient, and more effective, but the path was anything but. She realized she wasn't just implementing a tool. She was trying to change how an entire group of legal professionals thought about the way they worked. And, as she puts it plainly: "Not a single person wanted an intake system."

To combat this resistance, Joy didn't go looking for a better technology. Instead, she went looking for a better approach. A series of deliberate, human-centered steps that brought people along the journey rather than dragging them forward.

This resulted in a legal department that went from universal resistance to 90% adoption. And lawyers who once rolled their eyes at the idea of an intake system were asking, "Is Ask Al back up yet?"

So what does it actually take to get there? It starts with asking a different question altogether.

Start With the Problem, Not the Product

There's a pattern that plays out in legal transformation efforts more often than anyone would like to admit. A senior leader comes back from a conference, or reads an article, or sits through a vendor demo, and the directive comes down: "We need to get AI."

To do what, exactly?

That question — deceptively simple, surprisingly hard to answer — is where most transformation efforts go wrong before they've even begun. Technology gets procured, rolled out, and then eventually abandoned because nobody took the time to connect it to an actual problem that actual people needed solved.
‍

‍

At Altria, the answer to that question required real legwork. A consultant came in and mapped every single process across every practice group. They found that redundant, repetitive, transactional tasks are happening in silos across the department. This is work that nobody had stepped back to look at in aggregate and pain points had become so normalized that people had stopped recognizing them as problems at all.

So, when you ask people directly what their pain points are, you often don't get the full picture as it's not an easy question for people to answer. The better approach is to sit in the rooms where the work actually happens, listen carefully, and let the problems reveal themselves. Sometimes the most valuable thing a legal ops professional can do is put down the survey and pick up a chair.

Only once you have genuine clarity on the problem does the question of technology become useful. And when you approach it that way, the answer might surprise you. It might be a purpose-built tool. It might be something you already have. It might be as simple as a Power Automate workflow. The point is that the problem leads you to the solution.

Buy the answer, not the product.

Perfect Is the Enemy of Launched

Here's a scenario that will be familiar to anyone who has worked in legal operations. 

The team identifies a genuine problem. They agree a solution is needed. They begin scoping what that solution should look like. And then, slowly, the project starts to expand. More requirements get added, more stakeholders get consulted, and more edge cases get accounted for. Six months later, nothing has shipped, the momentum has gone, and the original problem is still very much unsolved.

Joy knew this going in. With a hard go-live date and a department that was already sceptical, she made a deliberate choice to launch something imperfect rather than wait for something polished. 
‍

‍

That transparency was itself a strategy. By naming the limitations upfront, Joy took the weapon out of the hands of the sceptics. Nobody could point at a gap in the system and declare it a failure, because she had already acknowledged that gaps existed and committed to closing them. The message to the department was clear: this is a foundation, not a finished product, and your feedback will shape what comes next.

Altria’s Solution: the AI Legal Front Door

Because the team was disciplined about what mattered most in phase one and ruthless about deferring everything else, Altria ended up with a Legal Front Door that went live in 52 days. 

They started with a small working group, a focused set of use cases, and a web form that captured the essentials. This was enough to prove the concept and build the habit.

What followed — the AI agent, the Teams integration, the document repository — came later, once the foundation had been tested and the resistance had begun to soften. Phase two was better because phase one had happened first.

There's a broader principle here that applies well beyond legal operations. In any change effort, early momentum is fragile and precious. Waiting until you have the complete solution means waiting until the energy has gone. A minimum viable product, launched with honesty and iterated with purpose, will outperform a perfect product that never ships every single time.

Meet People Where They Are, Or They Won't Come

When rolling out new technology, most people assume that if the tool is good enough, users will adapt to it and that the quality of the solution will be sufficient motivation to change behaviour. Essentially, users will meet the system halfway.

However, they won't. At least not reliably. Not at first. And in a legal department, where professionals are already stretched, sceptical, and convinced that the old way works well enough, the bar for friction is even lower than you think.

This was one of the earliest lessons Joy learned. When she first introduced the idea of a legal intake system, the feedback was immediate and unambiguous. Lawyers who were used to picking up the phone, sending a quick email, or firing off a Teams message wanted to know why they should change. And when Joy told them the new system would meet them where they were — that they wouldn't have to work differently — and then launched a web form, the response was "Joy, you said there would be no change yet. I used to text someone for this, but now I have to go fill out a form."

Because of this, they decided to introduce Ask Al — an AI agent built specifically to close the gap between what people were being asked to do and how they already worked. Submit a request by email? Ask Al handles it. In a Teams meeting when a request comes in? Open the app and submit it without leaving the meeting. The information collected is identical to the web form, but the experience fits inside the workflows lawyers were already living in.
‍

‍

‍

The impact of this seemingly simple design decision is hard to overstate. Joy describes a scenario where a lawyer receives a request during a meeting, submits it through Ask Al on the spot, and has a response waiting before the meeting ends.

This is the principle that underpins successful adoption in any context: the easier you make it to do the right thing, the more people will do it. And every point of friction you remove is a reason to comply that you don't have to manufacture. 

How to Work With Resistance: Lessons From Altria

No matter how well-designed your tool is, how carefully you've mapped the pain points, or how thoughtfully you've built for the user's reality, there will be resistance. But it’s not a sign that you've done something wrong. Resistance is inevitable when you're asking people to change, as change is uncomfortable even when it's clearly better. So, since you can’t eliminate it, how do you work with it intelligently?

Joy walked into Altria's legal transformation thinking carefully about its shape. She developed a different strategy for each form it took and categorized her audience into three groups. 

1. The first were willing to give it a shot if it saved them time — her carrot group. 
2. The second would only engage if they were required to — her stick group. 
3. The third refused outright, regardless of the merits. 

Most change efforts exhaust themselves trying to convert the unconvertible, burning energy and credibility in the process. Joy's instinct was to start where the momentum was (carrot and stick groups), build visible success, and let that success do the persuading. If the refusers eventually came around — as many did — it would be because the evidence became impossible to ignore, not because they were argued into submission.

For the carrot group, the approach was nurturing. Training sessions open to everyone. Regular communications from Ask Al — tips, tricks, updates — that kept the tool visible and approachable. And crucially, success stories. Real examples of time saved, workload reduced, and requests handled faster than before. If it worked for someone they knew, maybe it could work for them.

For the stick group, something more decisive was needed. And this is where executive sponsorship moved from nice-to-have to non-negotiable. Altria's General Counsel stood up in an all-department meeting and made it known that the expectation going forward was that everyone would use the system, including himself. And he followed through. The effect was immediate. When the most senior person in the room is not only endorsing the tool but visibly using it, the calculus for the reluctant majority shifts entirely. Suddenly the question isn't whether this is worth engaging with, it's whether you want to be the person who isn't.

People take their cues from the environment around them (i.e. social proof, authority, peer behaviour). So, building those signals deliberately through success stories, visible leadership, and clear expectations creates the conditions in which the right choice is also the easy choice. The goal, ultimately, is a mindset shift so that instead of thinking about whether to use the system, people simply use it, the same way they'd reach for email or pick up a phone. That doesn't happen overnight. But it does happen, if you're patient, strategic, and willing to meet the business where they actually are.

Brand It Like You Mean It

Here's something that doesn't appear in most legal technology implementation guides: give your tool a name. One that means something to the people who are going to use it.

It may sound trivial, but based on experience, it isn't.

When Joy was explaining the new intake system to a colleague in IT, he listened to her description and said, "Oh, it's like an Ask Al for the legal department." Al, short for Altria. Joy's response was immediate: "Brilliant. I am going to steal that." A logo was created. It went on the front page of the legal portal. It appeared on every communication, every tips-and-tricks email, every update from the centre. And suddenly, the brand took on a life of its own.

Lawyers started using the name in conversation. Joy would pass colleagues in the hallway, they'd start to ask her something, and then catch themselves: "Oh wait, should I Ask Al?" The system had stopped being a system and started being a resource. Something with a personality and a purpose that people understood intuitively.

This matters more than it might seem, for a specific reason. Legal departments are full of professionals who are deeply resistant to feeling like they're being processed. The language of intake systems, ticketing tools, and workflow automation carries an institutional coldness that does nothing to soften the resistance Joy was already navigating. But Ask Al? That's something you consult. Something you turn to. Something that feels, at least a little, like it's on your side.

Branding, done well, does something that training and mandates cannot: it shifts the emotional register. It takes a tool that people are being asked to use and turns it into something they want to be associated with.

Steps to Branding Your Legal AI Tool

In a department full of smart, busy, sceptical people, out of sight really is out of mind. So, the antidote is a brand that earns its place in the conversation and keeps showing up in new and exciting ways.

What Good Looks Like: The Metrics That Actually Matter

There's a version of this story that ends at 90% adoption and calls it a win. And it is a win — a significant one, particularly given where things started. But adoption is a means to an end, not the end itself. The more important question is what that adoption makes possible that wasn't possible before.

For Joy, the most immediate and tangible benefit wasn't a metric at all. It was visibility. Before Ask Al, work lived in individual email inboxes, on personal to-do lists, in the memories of lawyers who were already juggling more than they could comfortably handle. There was no way to see, at a glance, what was in flight, who was working on what, or whether the workload was distributed in any rational way.

That visibility changes the nature of leadership in a legal operations context. When you can see the work, you can manage it. You can identify when someone is overwhelmed and redistribute. You can spot gaps in capacity before they become missed deadlines. You can answer the question — with data rather than instinct — of whether the team is appropriately resourced for the volume coming in. After a period of organizational restructuring that included reducing administrative headcount, this kind of oversight became not just useful but essential.

Additionally, the metrics you can report on in month seven are not the metrics you'll be reporting on in year two. The early wins — adoption rates, request volumes, workload visibility — are valuable in their own right, but they are also building the foundation for the more sophisticated analysis that comes later. Turnaround times, cost avoidance, reduction in outside counsel spend, and business satisfaction scores take time to accumulate and time to trust. Start collecting now.

Key Takeaways

Legal transformation is not a technology problem. Choosing the right tool, designing it carefully, and building it with the user's reality in mind does matter. But the technology is the easy part. 

What's hard is the human part. The change management, resistance, and internal politics of asking a group of highly trained, deeply independent professionals to do something differently. The patience required to build momentum slowly, iteratively, without the guarantee that it will work.

Altria's legal team achieved 90% adoption because they:

• Thought carefully about people before she thought about products. 
• Launched before she was ready and kept improving. 
• Met resistance with strategy rather than frustration. 
• Built a brand that lawyers actually wanted to be part of. 
• Had a General Counsel willing to stand up in a room and lead by example. 
• Consistently showed up, communicated, and made the case over months.

The broader lesson, for anyone navigating their own version of this journey, is deceptively simple. Start with the problem, launch something imperfect, design for the humans in front of you, accept that resistance is part of the process and work with it rather than against it, build the data infrastructure now, and don't wait until you have it all figured out.

If you're ready to take that first step, we'd love to help you figure out where to begin. Schedule a call with us today and let's talk about what the right starting point looks like for you.