Comment by wolfrom | Hacker News Reader

Comment by wolfrom | Hacker News Reader

Parent Full thread

The concept we are working from is creating an API and possibly an embedded JavaScript to analyze the data and to provide updates, alerts and reporting. The plan is to make integration for tracking users as simple as possible to implement, while allowing for heavier customization for alerts and feature control if desired.

Where we've started from is that we have a product that we're going to be releasing in the next few weeks, and we felt that not only did we need to develop beta/release functionality, but that we would need it for several successor products that we will be developing. Rather than creating a compiled component, we thought that an API-based tool would give us more flexibility in future platforms and in sharing the features with others.

Like all other delegation, if we can't provide more benefit than you'd be able to develop for an equal or lesser cost, we're not worth it. But from the work we've done so far, we believe that we can make the process more efficient for startups.

That may not be enough specific info for the moment, but we will be fleshing that out more in the coming weeks. Thank you very much for your feedback.

I appreciate that you may not be able to give full details yet. We're not advertising much about what we're doing either, until we're ready to ramp up the marketing properly as we approach launch time.

The one suggestion I would make for when you are able to reveal more is that when you write things like this:

> The concept we are working from is creating an API and possibly an embedded JavaScript to analyze the data and to provide updates, alerts and reporting.

it sounds like something I might potentially be interested in, but I'd need to know in concrete terms what the "data" here is.

I suspect we're not alone in planning a service that is, essentially, a way to access a database from a browser. Providing that access is the easy bit, relatively speaking. More difficult is collecting useful information to make decisions about how to grow the service, particularly in the early months when there are many ideas but not enough resources to chase all of them at once.

If you can make life easier in that department, I imagine we'd at least take a look. However, I have to say in fairness that with so many similar services around, we'd need to have a clear idea of what might be in it for us very quickly or we wouldn't spend any time investigating further. It's not as if any startup running a database-type service like ours isn't going to have plans for in-house analytics that are already tailored to their specific needs, so I think you've got a potentially useful idea but probably a tough sell until you can give more detail. Maybe this poll is a bit premature.

This is very good information for us. This poll is part of our "is this a marketable idea" phase. It will help us decide if we should focus only on these tools for our products, or if we should design it for use by other applications through the API.

I had my reservations about people being willing to rely on an external service for this type of functionality, but it sounds like the bigger concern is more of value and utility. This should help us work towards the right design.

FWIW, I personally have no trouble relying on external services, provided that they are reputable and unlikely to disappear or otherwise let us down. As we're in the UK and many such services are abroad, legal action is never likely to be relevant here, so trust is everything.

You are quite right that I am far more interested in the cost/benefit question. To give you a specific example, we could build a monthly credit card billing system ourselves, using nothing but a payment gateway and merchant account. However, we know this would be painful, since anything to do with handling credit cards in-house is painful anyway, and setting up a robust system to do that would take a huge amount of development time that we would rather invest in our service. Thus we are seriously evaluating a small number of automated billing services, and fully expect to use one or other of them provided we can set up the pricing plans we need on their system by launch time.

On the other hand, we are unlikely to use Google Analytics. Leaving aside the "It's Google"/privacy concerns, I just don't find it particularly powerful or easy to use. On the other other hand, a good analytics system for our purposes, which could be integrated with our back-end database code to get whatever source data we want to evaluate but handle all the maths/pretty pictures/etc. easily, would be of substantial value to us. I'm not sure whether this is the sort of thing you have in mind, which at this point is a problem for both of us. ;-)

Reply on news.ycombinator.com