How to spot things early with Google Trends

A few methods for getting more out of this incredible tool

Hey everyone,

Every founder or potential entrepreneur is looking for what could explode or become a big topic.

One approach for that is using Google Trends.

This is one of the best free research tools out there, invaluable in fact, for everything from SEO purposes to investing.

Google, is the world's biggest search engine with dominant market share across many countries... e.g. 93% share in the UK.

So it reflects what the general population is caring about, aware of, making intents to buy and so forth.

It doesn't capture everything e.g. sites getting tons of direct traffic or early stage startups. But it is a very good tool you can use completely for FREE as a business owner.

It solves the problem of knowing if there is DEMAND for something, if people are searching for something there is demand for it!

'Watch what people do, not what they say' and all that jazz.

Many people use it in sloppy ways for spotting new trends.

For example, they take the top 25 rising queries for a topic and scan through those, plotting them over time.

Time-consuming and not specific enough.

The fact is 90+% of these will be irrelevant, because they just have sudden spikes for near very low absolute volumes (monthly searches). In fact almost near zero in a lot of cases.

Hint, if you do this manual approach, rising queries over longer time periods will be better quality usually, but it's still not perfect.

There are layers to it and smarter ways to use the tool. Here are 3 kinds of ways you can use Google Trends to spot what might become a great industry or business.

1) Exponential pattern filters

One way I use Google Trends is using Python script with an exponential filter to find rising queries from different base keywords that we choose (over different time periods).

An example of the raw output is shown below. This was just as an example from a bunch of random supplement and beauty keywords.

Obvious stuff in the space is not interesting e.g. The Ordinary.

But other things might be e.g. 'pre stretched braiding hair' -> I know from another run of the algorithm that knotless braids have seen huge growth in search interest in the past 2-3 years or so.

Go look at the absolute volumes and see what has happened in the past 12 months. Go look at the first page of results in Google, is there a brand already there killing the space? Loyal customers/cult following, major YOY growth, tons of forum discussion, proven revenue or user numbers.

High volumes aren't necessarily bad or too competitive, it means there is lots of demand and likely pain points or gaps in the market that aren't being fully met by incumbent players.

The key, if there is a gap, is to then create a brand around that niche or passion point that audiences resonate with. E.g. if pre-stretched is an area you are interested in, don't create productname.com, create braidgang.com.

Or go broader if you know you are going to expand out of just braids in the long run, tie the name to black hair styling in general perhaps, pick an emotional connection. You get the idea. It's not my area so I'll leave the naming up to you (marleys.co.uk??)

As mentioned before, trends are more art than science.

The curve for 'free sugar daddy apps' above isn't particularly interesting, far too noisy. But Google web searching isn't necessarily where the app action happens.

What is happening with sugar daddy app downloads?

Go look at some in the Play Store, use App Annie and similar tools.

Has a Tinder of the sugar daddy space emerged yet? Go find out! What UX/UI issues are there at the moment with apps in the space?

Google Trends can act as a spark for new research you can go and conduct in an area you are interested in.

2) Seeing convergence over time

Another method is the 'convergence' approach. I.e. does a smaller trend start to converge on the amount of search interest for a bigger or established trend.

A quick example here is 'fat dissolving injections'...

They converged to a much larger % of liposuction interest and actually overtook fat freezing interest towards the end of 2019.

When the % of convergence to a larger trend tips over 5-10% and keeps growing, this is usually a good sign something is going to become a much bigger trend.

Of course, Google Trends is relative and indexed so you don't know the absolute volumes,

Always go check these in Keyword Planner or another keyword tool of your choice.

A pattern I like to see in the absolutes over the past 12 months is something continuing to double or exponentially increase its search volumes. Finding these at low levels is great for spotting things not on a lot of other people's radars.

E.g. 1,000 searches in January, 2,000 in March, 4,000 in May and so on.

3) Little early clues left

Sometimes, Google Trends doesn't show something that will be bigger, for whatever reason, maybe search isn't where it is blowing up.

But there are more often than not, little 'early clues'.

For example, if you see something in the top 25 rising queries for a niche or area you pay attention to avidly that you haven't seen before.

It could be a company getting early venture funding or their first piece of press.

If you check the company out and the offer looks compelling, they have influencers/celebrities raving about them organically, forum/Reddit interest etc. it is likely to get bigger and the search patterns will start showing up.

Empirical Spirits is one brand in this vein, no steady curve in Google Trends but organic influencer attention (Drinkhaus CEO) and early cult attention.

Also, things generating big search spikes without much ad/PR campaigns, this shows a team or topic able to generate virality which will likely get bigger.

Another kind of early clue for seasonal things is the start of a higher high/higher lows pattern.

I saw this when spotting Mirror Workout in early 2019 and it then went higher during Christmas 2019, even before lockdowns hit in the US it was showing this pattern. Veganism also exhibited this pattern, seeing higher highs and lows from 2014 or so onwards.

In Summary

In summary there are a few creative ways to look for something in Google Trends that might become big and can help you develop a new brand, learn some marketing tactics, introduce new products and even source investments.

• Exponential patterns at early stages. Programming can help you with this.

• Look for convergences on established trends. Can do this manually if you want.

• Look for early clues e.g. sudden appearance of something new in top 25 rising queries, that has other attention e.g. organic influencer mentions or forum discussion.

• Viral hits without big ad spend behind them.

• Pattern testing higher highs and higher lows for seasonal/event queries.

Any questions, feel free to ping me! I'm waiting.