Many companies plan on a market forecast they buy from an outside provider. It is a fast way to get a number, and for broad economic outlooks it is often good enough. The trouble starts when the forecast has to match your own markets and segments, and someone asks why the number changed.
Indicio is one alternative: a no-code platform that builds a forecast of your own market from leading indicators, and shows what drives it. This article compares the options, including the cases where buying still makes sense.
What a bought forecast gives you, and what it does not
- It is built for everyone. The provider forecasts a country or an industry. Your market is a slice of it, defined by your products, your customers and your regions.
- Your competitors have the same number. A forecast that anyone can buy does not help you see a turn before others do.
- It arrives on the provider’s schedule. You get an update each quarter or each month, not when the data changes.
- You cannot see inside it. When management asks why the outlook was cut, the honest answer is often “the provider revised it”.
- You cannot run your own scenarios. If you want to know what a rate cut or a tariff does to your market, you depend on what the provider chooses to publish.
Four alternatives
| Option | What you get | Where it falls short |
|---|---|---|
| Keep buying | A ready number with little effort | Generic, the same as competitors have, hard to explain |
| A consultancy builds a model | A forecast fitted to your market | It ages after delivery, and updates cost extra |
| An in-house data science team | Full control over data and method | Needs scarce specialists, and depends on them staying |
| A forecasting platform with leading indicators | Your own forecast, updated automatically and explained | The team has to learn a new way of working |
How to build your own forecast from leading indicators
- Define what you forecast. Use your own market definition: registrations, orders or volumes by country and segment.
- Find the indicators that lead it. Leading indicators, such as new orders, building permits or commodity prices, move before your market does. Indicio searches more than 400 million external indicators and keeps the ones that lead your series. See the guide to variable selection.
- Test many models. No single model is best for every market. More than 80 are backtested on your data and the best are weighted into one forecast.
- Compare with the forecast you buy. A backtest shows how each would have performed in past years. This is the fairest way to decide.
- Run scenarios. Set a path for an indicator and see the effect on your market. Read how scenarios work.
An example: one method for more than 100 countries
A market-leading maker of heavy-duty trucks needed market forecasts for more than 100 countries, with a small team. Each market had been forecast by individual judgement. With Indicio, leading indicators are selected automatically for each country, and every market follows the same structured process. The result was lower forecast error, earlier detection of market turns and a process that management backs. Read the heavy-duty trucks case.
When buying still makes sense
- You need a broad economic outlook, not a forecast of your own market.
- The market is small for you, and a rough number is enough.
- You want an outside view to compare with. A bought forecast is a useful benchmark for your own, and its assumptions can be used as a scenario.
The two can work together: your own forecast for the markets that decide your plan, and a bought outlook as a reference.
Frequently asked questions
What is a leading indicator?
A leading indicator is a data series that tends to move before the variable you want to forecast. New orders, building permits and business confidence are common examples. A model that uses them can see a market turn earlier than one that only uses the market’s own history.
Is it more accurate to build your own market forecast than to buy one?
It depends on the market, which is why a backtest matters. Building your own lets the forecast follow your market definition and the indicators that lead it, and a backtest against the bought forecast shows which performed better in past years.
Do you need data scientists to build your own forecast?
Not with a no-code platform. The platform handles indicator selection, model testing and combination. The team needs to know its market and be able to judge whether the drivers make sense.
