Is doing nothing really an option?
Yes, and it is worth treating as one rather than as a failure of nerve. Every firm has limited partner attention. Spending it on AI means not spending it on something else, and for some firms, at some moments, the something else matters more. The mistake is not waiting. The mistake is waiting without knowing what the wait costs.
That cost is not dramatic. It is the same hours going into the same write-ups, proposals and evidence packs this week as last week, paid for at senior charge-out rates and recorded nowhere. Because it does not grow suddenly, it is easy to leave alone.
How do the two paths compare side by side?
| Wait and watch | Act on one job now | |
|---|---|---|
| Cost structure | No new spend. The existing cost of the job continues every week. | A defined spend on one process, set against the hours it returns. |
| Time to value | None, by design. | Weeks, if the scope is one clearly named job. |
| Risk | Low visible risk. Slow risks: senior burnout, staff using unapproved tools, knowledge held in a few heads. | The rebuild misses the mark. Reduced if the fee is fixed and depends on go-live. |
| Management attention | Freed for other priorities. | Needed for a few weeks, mostly in the mapping stage. |
| Control over AI use | Limited. People experiment on their own. | Higher. One sanctioned process, with review built in. |
| What happens after | The same decision, a year later, with the same job still costing the same hours. | One process running, and evidence to decide on the next. |
When is waiting the right call?
When the firm is mid-change. If you are merging, moving practice management system, restructuring teams or replacing partners, the process you would rebuild may not exist in its current form in six months. Wait until it settles. Do not hire us, or anyone, to rebuild something that is about to change.
Waiting also makes sense when you genuinely cannot name a job that absorbs senior time. Some small, specialist firms do almost entirely bespoke work, with little that repeats. For them the gain from rebuilding any one process may be too small to justify the effort. Good habits and a sensible AI policy may be all that is needed for now.
What does waiting quietly cost?
Three things build up while nothing changes.
- Hours. The job keeps taking what it takes. To see the figure for your firm, multiply the weekly hours by the number of fee earners, by your charge-out rate and by 46 working weeks. The unbillable hours calculator does the sum.
- Unmanaged use. Staff who are not given a sanctioned way to use AI tend to find their own, often on client material. Shadow AI looks at what that means for a firm.
- Knowledge held by a few people. A job that lives in the heads of two senior people is fragile. When one leaves, the firm learns how much it depended on them. When knowledge leaves covers this.
How should you decide?
Do the one thing that costs nothing: measure. Pick the job your senior people complain about most and count the hours it takes in a normal week. Then put a value on those hours at your own rates.
If the figure is small, or the firm is about to change shape, wait and check again in six months. If the figure is large and the job is stable, waiting is a decision to keep paying it. The audit does the measuring for you in about three minutes and names the process to tackle first. Aldbry rebuilds that one job in 30 days for a fixed fee, and if it is not live in that time, you do not pay.