Handling the unmeasurable

There is docs value that's almost impossible to measure. Take Stripe's famous developer docs as an example. They have long been pointed to as an industry leader in developer docs experience. Some of the benefits from this will be measurable, but some are much trickier. They:

• Boost the overall product reputation.
• Encourage adoption through developer confidence and enthusiasm.
• Produce good word of mouth.

How do you measure that? No-one thinks they're not valuable aspects of the docs, but how do you make a case for working on them, or show that you've succeeded?

If you were a well-funded marketing team at a big brand, you could use brand tracking services. But no docs team is getting the funding for that sort of research (most marketing teams aren't either).

There are three approaches you can take:

Option 1: Only focus on the quantifiably measurable

And hope this leads to good effects on the unmeasurable side as well. The downside to this is that the metrics that are easy to get are also likely to focus on the 'short': current users, and potential customers who are already close to buying. Reputation and brand are 'long'. Three years from now, someone on Reddit is going to ask for tool recommendations. You want someone else to reply "Use Z, they're super developer friendly, their docs are amazing, here's their getting started guide". Neither of those people are anywhere in your metrics yet.

Option 2: Include qualitative data

User interviews and annecdotes need to be handled with care (one loud user certainly shouldn't drive your strategy) but they can give some indication of your documentation reputation.

A docs NPS score could be part of this.

Option 3: Allow time for gut feeling work

Allow your tech writers and devs the time to work on the things they believe will have long term benefits, and don't demand instant returns or hard numbers. This means accepting that you are gambling a percentage of your team's time. But if you've hired smart, motivated people, there's a high chance it'll pay off. This option has additional benefits in team happiness.

Recommendation

Combing option 2 and option 3. Accept that you're not going to have a very clear picture here, and that payoff may take time (which is why it's harder to make the case for this at early stage startups).