Stick to the plan, or pivot? Seize a new opportunity, or blink? Hold to the course or stride off in a bold new direction? These are the decisions young companies confront frequently. Which path should your business choose?

Related: 10 Behaviors of Unstoppable Entrepreneurs

Well, that depends: On one hand, strategy takes time to work. A mediocre strategy executed consistently over time may well do better than a series of brilliant strategies never fully executed.

On the other hand, there's no virtue in sticking to a plan merely because it's "the plan." Many startups have to pivot to succeed. So, there is the dilemma.

Now, here is my five-step plan to figure out when to pivot and when to stand your ground.

1. Start with concrete specifics you can track.

An old saying points out that, "The difference between a goal and a dream is a date." Specific, finite and measurable goals you can easily track make your progress easy to see, so don't hesitate to "drill down" when you need to. Big concepts are great, but they work better when they're reduced to solid numbers. You can't track a concept, but you can track pluses and minuses.

Build your plan on trackable milestones: sales numbers, cash flow, appointments, prospects, products, patents. . . Whatever you deal in, make sure you're dealing in reality. If your specialty is customer support, measure it in incidents, calls, minutes per call, whatever. If it's marketing, measure it in web visits, clicks and links-in. For social media, measure what you put into it (tweets, posts, updates) and what comes back (retweets, likes, shares).

Over time, you'll get multiple benefits from the right kind of plan. You'll accumulate a collection of milestones met. And the people in your management team will see their progress in terms of objective numbers, which usually feels better than the subjective pat on the back.

2. Keep existing assumptions in mind, but keep them up to date.

Plan around assumptions that connect your concrete specifics with the reasoning behind them. For example, most online marketing has assumptions that say that how much you spend to generate traffic will produce how many visits you'll see coming to your website -- and how many orders.

Similarly, the sales forecast for a restaurant assumes meals per hour or diners per table; and the sales predictions for bicycles assume something about weather. Milestones for launch dates or promotions might assume attendance at a certain trade show. And marketing to generate leads might assume a certain volume of responses to email or of attendees turning up for a webinar. Make your own assumptions explicit so you can review them later.

Related: A Foolproof Method to Achieving Targets

3. Track and review your progress and your results.

Set a regular review schedule and keep to it. I worked for years with a plan-versus-actual-reviews meeting on the third Thursday of every month, so we could close the books on the previous month before we had the review. You can schedule such meetings in advance, so your team members know the schedule well and can plan their own travel and appointments around it.

Keep your meeting times short and efficient. We brought in lunch and were almost always done by 2 pm.

A review meeting should start with a review of assumptions, move to the plan versus the actual numbers, then examine: the results of recent events, the milestones met and the milestones coming up. The more measurable the specifics, the more visible the good or bad performance will be.

Of course, plans are often wrong and things are not as expected, so the key decision for most meetings will be to what extent to revise and correct the plan. Those decisions have a lot to do with what makes a business successful.

4. Identify any changes in assumptions.

There's a good reason to have your next meeting start with a review of your key assumptions. Identify whether or not they've changed. When assumptions have changed, there is no virtue whatsoever in sticking to the plan you built on top of them. Use your common sense. Were you wrong about the whole thing, or just about the timing? Has something else happened, like market problems or disruptive technology or competition, to change your assumptions?

5. Evaluate execution.

At this point, you know what's been working and what hasn't; and you have a good guess about how your original assumptions have changed. Use your judgment to look at the most important differences between the plan and the actual results, to highlight the execution. Did things not work because the assumptions were wrong, or because the plan wasn't executed? Did sales fall behind because of poor selling, poor marketing or unexpected market problems?

And for the things that did work, for the good news that was received: Was the reason for the good news some positive change in the environment, or great execution? If sales were better than expected on some line of business, perhaps you put more resources into the marketing. Be sure to track everything.

So. . . stick to the plan, or change It?

Do not revise your plan glibly. Remember that some of the best strategies take longer to implement. Remember also that you're living with the plan every day; it is naturally going to seem old to you, even boring, long before the target audience gets it. There are no best practices or easy formulas to apply. You make these decisions based on the specifics.

But do it well, and your business will benefit.

Related: Are You Prepared for That Cliff That Could Be Further Down the Road?