vovonep

Hello Sebastian,

Can you please explain your statement a bit more in depth about that it’s actually cheaper now, compared to the $48 per year and $35 per year for the additional agents.

Please understand that we came on board to test your software with the $48 per year, for two agents, and $2 dollar per month for every next agent around the 20th of april. 2013. a few weeks later is was $48 for the first agent and $3 dollar per month for every next agents, that was about a week ago.

Now you are saying that even WITH discount, the price is $6 per month. I don’t understand how it can be cheaper now?

You are also comparing it with features, but the chat worked great the way we tested. Simple, relative lightweight, plain, the way we wanted it, and you told me even that that was the way FlexyTalk should be in your opinion, being proud of the lightweight backend.

Now you say you added features, which we don’t need, and since the client responds way slower right now (and wrong even due bugs), not really matching your earlier philosophy about the client.

But maybe i will understand when you specify which features are added to justify the steep price raise, and why we should benefit from them, so I can explain this internally and see if this is still the best option for our own purpose, but also the several customers that we advised we found and are testing a nice lightweight but mostly affordable solution for their (small) websites.

Thanks

Hi Sebastian,

I call it also lightweight, since we _did_ test many other clients and solutions! What I tried to achieve is to make it ‘even more’ (really) lightweight 😉

Please remind I’m not referring to ‘how fast the page loads’ with the tool link I gave you. It’s just a tool that shows nicely how many requests are made when loading a page, aka how ‘heavy’ a page is. What I tried to show is that our ‘full-blown’ page has 48 Requests without FlexyTalk activated, achieved by combining CSS, combining Java, using Sprite Images, lazy loads, etc., all to make the elements/requests that are initially loaded on each page (header/footer, etc.) are as low as possible.

With the FlexyTalk client activated, the requests raise from 48 to 60, which is 12 requests more for just one added functionality. And looking at the elements loaded for FlexyTalk, this can be optimized with some widely available techniques.

One of the things that you could do, is make a lazy load, starting with only loading the bare needed elements (preferably combined and with sprite images) to only show the ‘Chat/3D Icon’, and lazy-load all the functionality needed to connect to the backend as soon as you ‘mouse over’ the ‘Click To Chat’ element. There will nothing change functional really, since you don’t change the Icon/Chat part right now ‘live’ to show ‘agent’ availability upfront. (Which is a good thing is this case).

Look for instance @ our Facebook/Twitter/Google+ buttons below each Page/Post. Those things are also quite ‘heavy’ calls when you embed them into your pages, raising the page elements and (API) requests a lot, while many users don’t care about these buttons and don’t use them. Right now they are ‘static icons’ that gets activate on mouse over, so only then all the FB, Twitter, Google api calls/requests get executed.

It’s not nitpicking I’m doing. We just like to comply with the ‘Page Speed’ initiatives where for instance Google is part of too. Making internet connections faster is one thing, making Web Applications/Pages ‘more economic’ is the other thing!

And also think of this: Making the plugin ‘lazy load’ will also reduce the calls to your API servers significant, eliminating all the call from people’s page-loads that doesn’t want to use the chat, so it is a win-win situation.

Bottom line, to make it clear; I’m convinced the FlexyTalk is lightweight in perspective you refer to, but I hope you understand my reasoning too 🙂

If you want to discuss this outside the forums, please send me a PM, we can maybe help you with some ideas and experiences, testing and even code contributions.

– Casper