Problem Trend Analysis SOP

A problem trend analysis SOP tells you whether the problem you want to solve is growing, flat or fading, and why. It takes three to five days, plus half an hour a month to keep watching. Timing decides as many startups as product quality does. A real problem that is shrinking makes every sale harder each year, and a growing one pulls customers toward you before you spend on marketing.

What is the problem trend analysis SOP and when should you use it?

Problem trend analysis tracks the direction of a problem over time using several independent signals, then explains the change with named drivers. It answers the "why now" question that Sequoia Capital's well-known pitch outline asks every founder to address: what changed recently that makes this problem bigger or newly solvable?

The method relies on triangulation. Google Trends alone can mislead, because it shows interest relative to its own peak (scaled 0 to 100) and reacts to news, memes and seasons. When search, community activity, hiring and regulation all point the same way, you can trust the direction. It also uses the idea behind Gartner's Hype Cycle: new topics spike, crash and only then settle into steady use, so a spike alone proves nothing.

Use this SOP:

• When your idea rides a shift: new regulation, a new technology, a change in how people work or live.
• When an investor asks "why now?" and you want an answer with evidence.
• When search volume looks modest and you want to know if it is the start of something bigger or the tail of something old.

Scope check: this SOP measures direction and drivers. For the size of current demand, use the problem search volume analysis SOP.

What do you need before you start?

• A validated problem statement and the problem keywords from your search volume work.
• Two control terms: stable, related searches to compare against, such as a broad category word.
• Access to signal sources: Google Trends (free), a community where your customers talk, a job site such as LinkedIn Jobs or your national job board, and news search.
• A spreadsheet with one tab per signal and a monthly date column.

How do you run it, step by step?

1. Choose 5 to 10 tracking terms

Pick the problem phrases with the clearest intent from your keyword cluster, plus your two control terms. In Google Trends, compare up to five terms at a time with the same control term in every comparison, so the scales line up.

2. Pull five years of search interest

Set Google Trends to the past five years and your target country. Export the data. Mark seasonal patterns (tax deadlines, school terms, holidays) and compare January to January to remove them. Write down the direction: up, flat or down, and by roughly how much.

3. Add three non-search signals

Community: count monthly posts or new members in two or three relevant subreddits, forums or groups. Hiring: count job postings that mention the problem or a role that handles it. News and regulation: list articles, laws, standards or platform changes tied to the problem, with dates. Exploding Topics and Glimpse add growth labels for newer topics if you have budget.

4. Name the drivers

For every upward or downward move, ask what caused it. Sort causes with the PESTLE checklist: political, economic, social, technological, legal, environmental. A driver with a date and a source (a law taking effect, a platform policy change) is stronger than a general mood.

5. Separate lasting shifts from hype

Look at every spike. If interest returns to its old baseline within a few months, treat it as hype. If it settles at a new, higher level, treat it as a step change. Check where your problem sits on the Hype Cycle pattern: rising, crashing or settled.

6. Score direction and durability

Give each signal a direction (up, flat, down). Give each driver a durability score from 1 to 5: will it still be there in three years? A law with a fixed start date scores high. A viral video scores low.

7. Write the why-now paragraph and set up monitoring

Write three sentences: what changed, when, and why it makes the problem bigger now. Set Google Alerts for your drivers and add a monthly 30-minute slot to update the spreadsheet.

Template you can copy

TREND TRACKER (one tab per signal, one row per month)
Month | Term/source | Value | Note (event, season, news)

SIGNAL SUMMARY
Signal            | 5-year direction | Change     | Source
Search interest   | up / flat / down | __% (Jan vs Jan) | Google Trends
Community activity| up / flat / down | __ posts/month   | r/____, forum ____
Job postings      | up / flat / down | __ postings      | LinkedIn Jobs / job board
News & regulation | up / flat / down | __ items/year    | news search, official sites

DRIVERS (PESTLE)
Driver | Category | Date | Source | Durability 1–5 | Effect on problem

WHY NOW (three sentences)
"Since [date], [driver] has [effect]. Because of that, [customer] now
[problem consequence]. This will continue because [durable reason]."

Worked example: remote team managers and time-zone scheduling

This is an illustrative case with invented numbers to show the method. A founder wants to help managers of remote teams spread across many time zones plan meetings and handovers.

Search interest for her five problem phrases rises from an average index of 22 to 61 over five years, while her control term stays flat. The spike in one early spring triples interest for three months, then falls back, so she treats it as hype. The steady rise after it holds for three years, which she treats as a step change.

Community posts about time-zone scheduling in two remote-work forums grow from about 40 to about 110 a month. Job postings that mention "distributed team" or "async" grow too. Her news column shows several companies announcing permanent remote hiring across countries, and that is the driver she scores highest, at 4 of 5.

Four of four signals point up. Her why-now paragraph cites cross-border hiring as the driver, and she moves on to market sizing with a growing problem behind her.

What should you measure, and when do you stop?

Stop when every signal has five years of data and every move has a named driver or a "no clear cause" note. A declining problem can still support a business if you serve the remaining customers well, but expect costs per customer to rise each year.

Which mistakes should you avoid?

• Reading Google Trends as volume. It shows relative interest, scaled to its own peak. Use it for direction, and use keyword tools for size.
• Trusting one signal. Search rises with news coverage. Confirm with communities and hiring.
• Calling a spike a trend. Wait for the post-spike level before you decide.
• Ignoring seasons. A December dip in B2B searches comes from holidays, so compare December with December.
• Skipping the driver. A rising line without a cause gives you no idea when it will stop.

How can AI speed this up?

An AI assistant can suggest tracking terms and control terms, list PESTLE drivers for your industry and draft the why-now paragraph. Paste exported Trends data and ask it to remove seasonality and describe the direction. Check every driver it names against a dated source, since AI mixes up years and invents regulations. For pitching the result, the F/MS investor playbook shows where a why-now slide fits, and the F/MS startup playbook blog has more on reading signals before you build.

Where does this SOP fit in your validation plan?

Run it right after problem search volume analysis, using the same keywords. Social media listening and community research feed its community signal. Afterwards, the why-now paragraph goes into market sizing, market entry strategy validation and your fundraising validation. All ten SOPs of this stage are on the Problem Validation page.

FAQ

How do I know if a problem is growing?

Check four signals over five years: search interest, community activity, job postings and news or regulation. If three or more point up and you can name a lasting driver, the problem is growing.

Is Google Trends accurate enough for startup validation?

For direction, yes. It shows interest relative to its own peak, scaled 0 to 100, so it cannot tell you how many people search. Pair it with a keyword tool for volume.

What is a "why now" for a startup?

A short explanation of what changed recently that makes the problem bigger or newly solvable. Investors ask for it because timing decides whether customers are ready to buy.

How far back should a trend analysis go?

Five years for most problems. Go back further only if your problem moves slowly, such as changes tied to ageing populations or long regulation cycles.