Maximizing web app's business value through backend revamp | How we do IT #5

00:00 → 00:01

When it comes to web applications,

00:01 → 00:04

people tend to judge the book by its cover.

00:04 → 00:06

If it's nice and clean on the outside,

00:06 → 00:08

they are keen to use it.

00:08 → 00:11

However, it's on the inside that all the magic happens.

00:11 → 00:15

So today, we'll show you how to marry business logic with technology

00:15 → 00:19

to maximize applications' usability, especially if it

00:19 → 00:21

leaves a lot to be desired until now.

00:25 → 00:26

I'm Aleksandra,

00:26 → 00:28

Senior Content Marketing Specialist at Merixstudio,

00:28 → 00:30

and this is how we do it.

00:30 → 00:34

In this series, we present you specific design and development

00:34 → 00:38

challenges and explain how we deal with them at Merixstudio.

00:38 → 00:41

Every case is based on our own experience and projects we've

00:41 → 00:42

worked on so far.

00:42 → 00:44

In this episode,

00:44 → 00:47

we'll take a closer look at our client from the solar energy

00:47 → 00:50

sector who's based in Western Europe.

00:50 → 00:53

Their application is supposed to help salespeople prepare the

00:53 → 00:57

most accurate calculations and proposals for their customers.

00:57 → 00:59

The product has a lot to offer.

00:59 → 01:02

It estimates energy production and consumption,

01:02 → 01:06

visualizes data and prepares detailed offers for the customers.

01:06 → 01:09

It all sounds great in theory, but in practice,

01:09 → 01:12

the amount of actions puts a burden on the application,

01:12 → 01:15

causing it to work slowly and ineffectively.

01:15 → 01:18

How to make the most of the existing solutions while adding

01:18 → 01:19

value for the stakeholders?

01:19 → 01:22

See how we do it at Merixstudio.

01:22 → 01:26

Let's start with

01:26 → 01:27

the biggest challenges.

01:27 → 01:31

Firstly, the overall experience of the product is rather harsh.

01:31 → 01:34

The abundance of features can empower salespeople,

01:34 → 01:37

but accessing them takes way more time and effort

01:37 → 01:39

than it should.

01:39 → 01:43

Then, onboarding new employees to the tool can be tricky.

01:43 → 01:44

On top of that,

01:44 → 01:47

the company has grown and evolved and so the solution

01:47 → 01:49

stopped meeting users' expectations.

01:49 → 01:52

Sounds like a lot? Well, there's more.

01:52 → 01:55

The second issue is the tool's performance.

01:55 → 01:56

At the beginning,

01:56 → 02:00

the product was basically a spreadsheet with a complex algorithm.

02:00 → 02:04

Simply put, its function was to ensure the precision of the calculations.

02:04 → 02:07

As the tool contains a massive amount of data,

02:07 → 02:09

processing it takes a long time.

02:09 → 02:12

What's more, users access the solution via SharePoint,

02:12 → 02:15

so the environment is not entirely stable.

02:15 → 02:17

Not to mention that a spreadsheet doesn't really

02:17 → 02:19

facilitate the entire selling process.

02:19 → 02:21

Now that you know the problems,

02:21 → 02:23

it's time to learn about the solutions.

02:23 → 02:27

My expert today is Bartek, our Python developer.

02:27 → 02:31

So, Bartek, how about the high level changes in the product?

02:34 → 02:36

Improvement of a product like that should be conducted on a

02:36 → 02:40

few levels as we need to take care of more than one thing at a time.

02:40 → 02:44

Looks are not everything, but in this particular case, they do matter.

02:44 → 02:47

That's why the design work comes first.

02:47 → 02:48

Our app is all about the data.

02:48 → 02:52

So we need to find a way to visualize it properly.

02:52 → 02:55

Almost ninety screens and dozens of video calls with the client later,

02:55 → 03:00

we managed to achieve a cohesive light looking and fast working facade.

03:00 → 03:02

Remember, you make a first impression once,

03:02 → 03:04

so make it a good one.

03:04 → 03:06

Secondly, it's okay for the tool to be complex,

03:06 → 03:08

but not overwhelming.

03:08 → 03:10

An application of this kind needs a lot of features to

03:10 → 03:13

prepare a calculation or an offer for the customer.

03:13 → 03:17

But it doesn't mean all of them should be visible at first glance.

03:17 → 03:20

Plan tools look wisely, prioritizing the essentials

03:20 → 03:22

over the additions.

03:22 → 03:25

A well organized structure and navigation in application help

03:25 → 03:28

keep information architecture clean and consistent.

03:28 → 03:30

And lastly, think long term.

03:30 → 03:33

Because the company and the product will evolve.

03:33 → 03:35

Scalable design is one thing,

03:35 → 03:38

but scalable code is equally if not more important.

03:38 → 03:40

Following coding standards like KISS or YACNE

03:40 → 03:45

and writing clear acceptance criteria makes code lucid and concise.

03:45 → 03:48

It also lays the foundations for possible changes.

03:48 → 03:50

They will consume the new thing.

03:52 → 03:53

Now that you mentioned the code,

03:53 → 03:55

let me ask you about performance.

03:55 → 03:58

How would you boost it in this very case?

03:58 → 03:59

We've been there before,

03:59 → 04:01

so I have a couple of solutions to share.

04:01 → 04:03

To begin with, a spreadsheet may handle big data

04:03 → 04:05

calculations effectively.

04:05 → 04:09

But it's not exactly fit for creating user interfaces or

04:09 → 04:12

sales offers, especially in a growing organization.

04:12 → 04:16

That's why it has to be transformed into custom software.

04:16 → 04:18

Transferring the former solutions logic into it.

04:18 → 04:21

There are technologies that may easily handle the number of

04:21 → 04:25

calculations like Python connected with pandas or NumPy library.

04:25 → 04:26

And they may be a huge help here.

04:26 → 04:29

A clear organized code on the back end will increase the

04:29 → 04:31

application speed and stability.

04:31 → 04:35

Secondly, taking a deep dive into the product's business logic and

04:35 → 04:38

thinking about the best ways to translate it into something new

04:38 → 04:40

is a key in such optimization.

04:40 → 04:44

In this case, we deal with a very good algorithm inside the app.

04:44 → 04:45

We need to treat it respectfully.

04:45 → 04:48

Squeeze the best out of it and elevate it with some good

04:48 → 04:49

technological solutions.

04:49 → 04:53

That's the most effective way of realizing the full potential

04:53 → 04:54

of a product that needs a revamp.

04:54 → 04:56

Last, but not least,

04:56 → 04:59

integrations with external services may really make a difference.

04:59 → 05:03

In a former version of the app, all the data was hidden inside.

05:03 → 05:05

No wonder it tended to crash a lot.

05:05 → 05:08

It's always worth remembering that performance is crucial to

05:08 → 05:09

app success.

05:09 → 05:12

And think about the work optimization for the users.

05:12 → 05:15

Connecting the tool to the external solutions will

05:15 → 05:19

automatically perform actions that users would otherwise need to do manually.

05:19 → 05:20

Sounds like a win win situation.

05:20 → 05:21

Right?

05:21 → 05:23

Now, that's what I call useful advice.

05:23 → 05:24

Thank you, Bartek,

05:24 → 05:27

for walking us through complex back end matters in such an

05:27 → 05:28

approachable way.

05:28 → 05:31

Do you feel like your application could use a back end audit?

05:31 → 05:35

If so, follow our advice and let it shine brighter than ever.

05:35 → 05:39

Thanks for watching and see you in another how we do it episode.