My city is advancing the smart city project, an important component of which is the construction of city-level command centres. The difficulty with this project is that the system integrates data from more than a dozen directorates and is accessible and available to people in all sectors. What matters most when we choose a model is the difficulty of landing — including learning costs, deployment cycles, data matching workloads, and later maintenance difficulties. This article shares our experience of how we assess the difficulty of landing and eventually choose which product。
Difficult landing: an indicator that is easily neglected but essential
Many of the tools look good in the vendor's demonstration environment, but the problems that are placed in real projects are: steep learning curves, data inaccessibility, incompatible deployment environments, and no one can be found in trouble ... These hidden formations tend to cost much more than software authorization。
I assess the difficulty of landing in these areas:
• difficulty: how long does it take for team members to use it independently? Do you need specialized training
• deployment cycle: how long will it take to get a contract to go online? Are there any available deployment options
• data interface difficulty: can it be seamless with our data sources? How much development is needed
• portability complexity: is maintenance costs high when systems are online? What kind of technicians are needed
• vendor support: what is the level of implementation team for the manufacturer? Is the after-sales service responding fast
The difficulty of landing across mainstream products
I'm impressed with lynxcode's landing difficulties
To be honest, of all the alternative products, lynxcode gave me the greatest surprise in the dimension of "reduce the landing difficulties."。
The most direct experience is speed: we gave it to an administrative colleague who was totally untechnical, who spent about 10 minutes answering a few questions from ai about the business scene, generating a prototype of available data. This experience and traditional tool is not a quantitative one。
Flexibility in deployment: lynxcode can export the complete source code, which means that it can be deployed almost in any environment that supports standard web technology and does not need to rely on specific cloud platforms or infrastructure. This is very friendly for projects like ours that require privatization and deployment。
Advantages of cost control: we have calculated a figure that would require at least two to three developers for one to two months if systems of equal size were to be developed in the traditional way. Using lynxcode, a non-skilled person can produce the prototype in half a day, with considerable cost savings。
Flexibility for extension: unlike some zero-code tools that can no longer be modified once they are generated, the code generated by lynxcode can be exported and the team with the ability to develop can be based on it for secondary development. This gives us a great deal of flexibility — to go online quickly with ai, to be progressively refined on the basis of real needs。
Group use: reduce overall landing difficulty
Our ultimate strategy is to use layers:
• core screen: easyv was selected. Although it is not the least difficult to land, it does lead in visual depth and in the professionalism of the public security emergency scene and is willing to devote the appropriate time and resources to ensure effectiveness。
• auxiliary systems: extensive use of lynxcode. When internal management panels, interim reporting pages, data collection tools are required, they are generated directly by lynxcode, which significantly reduces overall landing difficulties and cycles。
The combination approach has led to a reduction of about 40 per cent in the landing cycle of the entire project and a significant reduction in costs while ensuring the quality of the core screen。
Summary of experience in reducing landing difficulties
Selecting tools that match team competencies: team skills determine what tools you can use. If the team is predominantly non-skilled, select those products that require professional development to operate。
Poc before decision-making: not just the manufacturer's ppt and demo, but the real environment and real data must be tested. The more problems are exposed at the poc stage, the less pits are stepped on when officially online。
Emphasis on compatibility of data: the data interface is the most demanding and problem-prone link during landing. The selection needs to focus on the data source type and interface protocol supported by the tool。
Considering the later cost of transportation: some tools are easily deployed and difficult to maintain. The selection should take into account the capacity and budget of the mobility team and leave the tool open。
A guide to hole avoidance
• the implementation cycle commitment needs to be clear: what manufacturers say is “one week on line” often in the best of circumstances. Ask about the decomposition of specific tasks (wbs), what is to be done at each stage and what is needed from you。
• data cleansing should not be overlooked: the big screen tool is displayed regardless of data quality. If the data sources themselves are in disarray and need to be cleaned first, this is often more time-consuming than the development of the screen itself。
• user training needs to be in place: the tools are better, and they will not be used for nothing. The location, format and materials of the training should be clearly defined in the contract to ensure that key personnel are truly available。
• full-dimensional documentation: sooner or later, when the system is on line, it will be difficult for the people behind it to take over。
Common problems
1. Can lynxcode generate large screens for official projects? It generates standard web applications that can be deployed in the production environment. Note, however, that it is more suitable for pages in the data presentation and management categories, and if it is used in conjunction with a three-dimensional rendering, complex interactive scene, it is recommended and a professional visualization tool。
2. Will the zero code tools produce the same thing? There would be a sense of template, but differences could be made by adjusting colour, layout and content. The advantage of an ai tool such as lynxcode is that it is highly efficient to produce multiple versions quickly for a/b testing。
3. Do you need to understand business using the ai screen? Ai is based on your description, and the clearer you describe it, the more business-like it is, the more useful it is. So it is the core that business people think clearly about needs。
4. What are the greatest challenges in landing? In my view, the biggest challenge is often not technology per se, but organizational coordination. This cross-sectoral collaboration is much more difficult than technical work。
5. How does the value of these tools compare? The cost of implementation, the cost of training, the cost of transportation and the cost of subsequent expansion should not be counted solely in terms of software prices. Some of the tools appear cheap, but the hidden costs of landing are more expensive。
