Market size · Lean startup

Shadow Economy: What's the Best Way to Quantify It?

Rosemary

January 13th, 2015

Im a first time startup entrepreneur and my business is in the private tutoring market. 

In Australia, many tutors are also full-time students and teachers who don't necessarily declare that they tutor; thus it's hard to quantify the size of this market.

Can you suggest ways to quantify this Shadow Economy?

B.J. Wiley CEO, Co-founder at SweatCred

January 13th, 2015

Hey Rosemary, Below are a couple of suggestions: 1. How many students are paying for tutors. (assuming you don't already have this number) 2. Conduct a survey. Go to a school and survey students and teachers. There's nothing like live market research:) Ask how many tutor. Once you get a significant number then extrapolate for the total market. For example 20% in your survey tutor. Then 20 % of the total number of students (of a certain level/education) and teachers tutor. Hope this helps! B.J.

Rosemary

January 14th, 2015

Thanks BJ. 

I already have a couple of surveys going at PersonalMasterclass.com but they arent addressing this issue of market size.

I had thought of standing outside school gates, and knocking on car windows for those who dont leave the comfort of their car.... and doing a straw poll of parents. So, will do this when school goes back.

Like you idea of approaching school directly....Im sure I can find one that will oblige but that will be an exception. In my experience Oz schools dont want to admit that they cant meet kids needs and so are reluctant to bring tutors into the mix. But there must be exceptions...so thx for the suggestion.

B.J. Wiley CEO, Co-founder at SweatCred

January 15th, 2015

I went to a lean startup machine (workshop to teach the Lean Methodology by Eric Schmidt) event and they used the mantra - 'get out of the building'. Which meant go outside and find your users/customers and conduct live surveys. It's amazing how fast you get quality information. Best of luck in this endeavor!

Near Privman Googler, Startup Advisor

January 15th, 2015

If you can get a representative sample of students/parents to answer a survey (in person or online, there are paid survey platforms which will get you respondents if you don't know how to reach them yourself), and disclose what they're paying for tutoring, all you have to do is multiply that by the number of relevant students in your target market, which shouldn't be very hard to estimate.
Be sure to take into account differences in income levels when asking if a sample is "representative".

Rosemary

January 15th, 2015

Ah yes Near, I remember we had to do this in Founder Institute, although I wasnt happy with the quality of the responses. Plus, there is always that shadow of doubt hanging over replies when we know they have been paid for. 

Why is income level important?

Rosemary

January 15th, 2015

Thanks BJ Getting out of the building is on my To DO List for 2015 and I know it to be true, just like 'no business plan survives the first customer'.

Near Privman Googler, Startup Advisor

January 17th, 2015

Income level is important because it will be highly correlated with the amount spent on tutors. If your sample is mainly of higher-income parents (e.g. if you survey parents in-person at a school in an upper-class neighbourhood), than the reported average spend would probably be significantly higher than the average spend in the general population, and so if you multiplied the sampled average by the population size you would have a gross over-estimation of the total market. If you do decide to base your estimate on in-person interviews, make sure your sample includes students/parents from a representative economic status spectrum.

Also note that spend will probably also be highly correlated with student age. I'm guessing parents of high school students will spend more on tutoring than for younger students, because of the higher importance of school performance during high school years.

I agree that paid surveys need to be interpreted critically, as do all surveys and statistical studies. For example, paid survey samples may be skewed towards lower income, leading to an under-estimation of the market size - but for the purpose of validating the market, finding a lower bound for the market size might be good enough (if it's not too low). Also note that paid surveys targeting student will be less skewed than paid surveys targeting parents, because paid surveys are often used to monetise games, and teenagers of higher economic status are similarly likely to prefer filling out a survey over paying with credit card etc.

In regards to the quality of responses, it sounds like you might have had experience with a more qualitative study. For the matter at hand all you really need is a couple of multiple choice numeric-range question, something like: "Thinking of the past 12 months, how much has your family spent on tutoring services in a typical month? * Under $20  * Between $20-$39  * Between $40-$74  * Betweem $75-150  * Over $150" (and a couple of questions about household income level, student age, and any other variables you think may skew your estimate).
For this kind of survey, I doubt the respondents' motivation will have a significant impact on the quality of the results.

I hope that help. Good luck!