I’ve learned a lot of things in my time at Atomic Object, most of them falling into two categories: how to write great software, and how to be a consultant.
One of the most important skills I’ve learned in the latter category is how to always have an answer for a client.
Consulting is expensive, and one of its main purposes is to answer client questions. These two things together mean that a simple “I don’t know” is rarely an acceptable answer.
So what do you do then when you really don’t know the answer? Here are a few techniques I’ve learned to give satisfactory answers to questions I don’t know, along with some guidelines to consider when using them.
1. Make an Educated Guess
Guessing the answer to a question can be dangerous, but often when the stakes are low and the cost of finding the real answer is high, this can be the right solution. In the software world, I often use this when working with client developers who want an opinion on how to code something. Researching the perfect answer isn’t generally worth it, and if my suggestion is bad, it usually only impacts a single story or requires a refactor down the road.
Things to consider
- Use this technique with caution. Only guess if there is a low negative cost for you being wrong.
- Always be explicit about the fact you are guessing. Never present a guess as fact.
- Tell your client why you are guessing, and offer to invest time in finding a concrete answer if they feel it is important. Be prepared with an estimate of how long it will take to find a concrete answer.
2. Direct Them to Someone Who Does Know
This is a simple redirection. If you know of someone likely able to answer a question that has you stumped, it should be your go to technique.
Make sure you don’t just brush your client off with an, “I don’t know go ask so-and-so,” or something similar. Doing so can communicate to the client that you don’t care to take the time to answer them or that you don’t know and don’t care to know what the answer is.
I usually tell the client that the other person would be better able to answer the question, then ask them myself either by walking over to their desk with the client or sending the other person an email and copying the client on it. This has several advantages:
- By asking the question yourself, you communicate to the client that you care about the question and answer too.
- You have an opportunity to expand your own knowledge by hearing what the answer to the question is.
- It puts you in a good position to continue helping your client find an answer if the other person doesn’t know either.
Things to consider
- Be prepared for the eventuality that the person you redirect to doesn’t know the answer. If that is the case, you may have to either direct your client to another person, or use another of the techniques below to find an answer.
- Be respectful of others’ time when involving more people in answering the question. If it seems like the other person will be able to answer in a matter of minutes, go for it. Depending on the urgency and gravity of the question, if it is taking a long time to get answers and several people are involved, it may make more sense to back off and try a different technique.
3. Offer to Pair on Finding a Solution
This is a solid technique for problems that require a “how-to” type answer, e.g. how do I make our framework do X. The big win is that both you and the person asking the question gain knowledge of not only the answer, but the path to get there.
Things to consider
- Be conscious of time. Don’t spend a large amount of time pairing on a small problem.
- Know your audience. Some people are willing to get down in the dirt with you and find an answer, and others expect you to do it yourself.
4. Offer to Research an Answer
This is a good technique for questions that can’t be effectively answered by any of the other techniques. Verbally committing to do research is generally enough to set your client’s mind at ease.
While it may seem like a small difference “I don’t know” and “I don’t know, but I will find out” are very different answers, and the latter will be much better received by most clients.
Things to consider
- Give a concrete time commitment. Instead of just saying, “I will look into it,” say, “I will look into later today when I finish my current task.”
- Give an estimate on how long you think research will take. If the number is too big, your client may decide other things are higher priority than getting an answer.
5. Identify Data Source(s) that May Provide the Answer
This is often a good preliminary step before jumping into technique 4. Instead of committing a large amount of time to researching an answer, identify what data you would need to obtain said answer, and communicate that to your client. This quickly gives the client the option of either gathering that data immediately, or delaying if the cost of gathering said data is too high right now.
In the software world, a good example of this is complicated UI decisions. A client may want an opinion on how to structure their UI, but often it can be beneficial to have them delay until they have data from things like usability test to make a decision.
Things to consider
- Be prepared for your client to push into technique 4 if they decide they really do need an answer right now.
- Try and have an idea of not only what data you need, but how to obtain said data.
Closing Thoughts
There will always be questions you don’t know the answer to, but just brushing them off with an “I don’t know” is bad customer service. Hopefully these techniques can help you always give your client a good solid answer or the means to get one.